JP2016540319A - Method and server for providing a set of quoted prices, eg airfare price quotes - Google Patents

Abstract

The technical field relates to methods, servers, and computer program products for providing a set of prices. The computer server, along with a request for prices for goods or services (eg airfare), parameters that define these goods or services, eg action types, eg airfare, hotel reservations, railway fares; date range; destination; Origin; desired weather conditions; star rating; keywords; receive any other user-defined preferences. Analyzing the pattern in that data set, one or more processors programmed with software then inferring, estimating or predicting the estimated price from an incomplete historical price data set And send the price quote to an end user computing device, such as a personal computer, smartphone, or tablet. [Selection] Figure 12

Description

  The technical field covers estimated prices, such as airfares, rail fares, hotel prices, goods (in fact, any type of goods or services, where the price is not fixed and is instead variable, Accordingly, it relates to a method, server, and computer program product for providing a set of quotes for prices for goods or services for which it is useful to be able to provide quotes for those prices.

  In the context of airfare, the price of airfare tickets for an itinerary is complex to calculate. The result of the calculation of possible routings, authorized fares and the availability of these fares. These calculations are typically performed in the Global Distribution System (GDS), and performing queries against such systems can be slow, expensive and performing calculations using energy. The result needs to be sent.

  Storing ticket prices and flight availability for all possible airports across all possible routes for a sufficiently long range (eg 1 year) requires very significant data storage capacity Would be.

  FR2841018 (B1) is a method for loading flight schedule changes, the step of receiving at least one set of flight changes, and retrieving individual changes in the set in a register for future scheduling. Storing, and simulating the reassignment of reservations related to flight schedule changes by accessing both the records and the flight schedule database via the reservation distribution server, and the reservation margin And finally updating the minute flight schedule database.

  As an example, in the context of airfare flight pricing, flight pricing is traditionally derived from GDS (how the GDS system works, and more generally how flight pricing works). (See Section C for a more detailed description of). Flight comparison services (eg Skyscanner) and some airlines pay a fee to get live bookable prices from GDS. These prices are the actual bookable prices that a prospective passenger can book. However, many prospective passengers often only browse and do not require a clearly accurate bookable price, but will instead be satisfied with a quote. As a result, it would be very useful if an accurate price estimation method and system could be devised, in particular it would not only be expensive to access third party resources (eg GDS), This is so if relying on access (which may not be available to obtain the necessary information) is removed.

  Thus, obtaining a price and flight availability estimate between any two airports over a period of time that is not slow enough to obtain the estimate by accessing the remote server or servers multiple times. It would be desirable to provide a method that does not require the enormous data storage capacity required to pre-store the results for all possible queries.

  According to a first aspect of the present invention, a method for providing a price quote, wherein (i) a computer server receives a request for a parameter defining an article or service along with a price for the article or service, such as an airfare. And (ii) determining an estimated price from an incomplete past price data set at any point in time for step (i) above, by analyzing the pattern in that data set, 1 Or (iii) configuring one or more processors to calculate an estimate for a requested price for an article or service that satisfies the parameters; and (iv) configuring the price estimate to an end user.・ Computing ・Vice, a method is provided comprising the steps of: sending example, personal computers, smartphones or tablet.

  One advantage is that this method requires the time or energy required to access one or more remote servers multiple times to actually set fare availability and calculate the fare price. It is not to do. A further advantage is that this method does not require the enormous data storage capacity required to pre-store the results for all possible queries. The computer performing the method operates in a new way and overcomes the recognized problem by the method. Note that steps (ii) and (iii) in this method are not necessarily separate steps, but instead may be part of the same step.

  The method includes parameters that define the goods or services that include action types such as airfare, hotel reservations, railway fares; date ranges; destinations; origins; desired weather conditions; star ratings; keywords; One or more user-defined preferences may be included.

  The method may determine the estimated price by inferring, deriving or predicting the estimated price.

  The method includes in step (ii): (a) obtaining past price estimates from a computer data store; (b) grouping past price estimates by category; and (c) for each group. Deriving statistical data; (d) storing a classifier including the derived statistical data for each group on the computer; and (e) using the stored classifier, the required price corresponds. Identifying a group may be included.

  The method uses step (iii) to calculate a set of estimates for the requested price over a specified date range using statistical data obtained from the stored classifier corresponding to the identified group. May be included.

  The method may include step (ii) comprising analyzing a pattern in the data set using rules.

  In the method, step (ii) may include a naive Bayes classifier machine learning approach that generates a probability model of the price, and the model may be used to predict an unobserved price.

  The method may train the classifier using one observed price and the corresponding set of features.

  Does this method feature on demand and does the departure date of the week, length of stay, travel involve spending a Saturday night at the destination? It may include one or more of airline, travel time, route, and month.

  The method would then allow the classifier to predict the price of an unobserved price by being given a set of features and giving the price most likely to have those features. May be.

  The method may derive the features by training a plurality of models with different features and comparing the prediction accuracy of different models.

  The method includes, in step (ii), building a statistical model from past prices, identifying missing estimation candidates, and pricing the estimation candidates based on the statistical model. It may be.

  The method includes estimating a price for each candidate estimate, extracting a category feature value from the estimate, extracting a trained classifier from the database for the extracted category, and an estimate candidate Extracting all feature values from, classifying candidate estimates, calculating the Bayesian posterior probability for each price range stored in the classifier, and the price range with the highest Bayesian posterior probability It may be performed in the steps performed by selecting a class and attaching a price class to a candidate estimate.

  The method may include an input to a statistical model including a list of paths, a classifier category classification scheme, a past estimate, and a supported set of features with weights.

  The method may filter past estimates by age.

  In the method, the input to the statistical model may include a backward path equivalent.

  The method may include the step of including the fare price cached in the set of price estimates.

  In this method, the set of price estimates may be configured without inquiring the distribution system after step (i).

  The method may be for a trip where the price is a one-way trip.

  The method may be for a trip whose price is a round trip.

  In the method, the price may include an air fare price.

  In this method, the price may include a rail fare price.

  In this method, the price may include a rental car price.

  In the method, the price may include a hotel price.

  In this method, the request may include a flexible search request.

  The method may be such that the final result of the method is at most one estimate per input date or input date pair.

  The method is as follows: (A) At any point in time for step (i), determining the confidence range of the estimated price from the incomplete past price data set by analyzing the pattern in that data set. Configuring one or more processors, and (B) configuring one or more processors to calculate an estimated price confidence range for a requested price for an article or service that satisfies the parameters. There may be.

  The method may further include the step of (C) providing a fare price estimate along with a confidence range to an end user computing device, such as a personal computer, smartphone, or tablet.

  The method may further include the step of determining, using the confidence range, whether to display the price to the user or provide the user with a likely price range.

  The method may display the likely price range as an error bar.

  The method may include only one departure date in the date range.

  In the method, the date range may include only one return date.

  The method may include a method that provides a set of best round trip price estimates for a specified departure date range and a specified return date range. Step (i) includes receiving a request for a specified departure date range for a trip from the origin to the destination and a round-trip fare price for the specified return date range, and step (iv) includes a specified departure date. It includes providing a set of best round-trip fare price estimates for a day range and a specified return date range.

  The method may be such that the best round-trip price quote set is provided in a bar chart format.

  The method uses a random variation that the estimation process adds to the minimum Bayesian posterior probability required to accept the classification result, the maximum number of path operators involved in generating candidates, or the Bayesian posterior probability to avoid ties. May be parameterized by one or more of them.

  The method may be performed on a server that provides a fare availability and price quote service.

  According to a second aspect of the present invention, a server configured to provide a price quote, receiving (i) a request for parameters defining the goods or services along with a price for the goods or services, eg airfare. Parameters include action type, eg air fare, hotel reservation, rail fare; date range; destination; origin; desired weather conditions; star rating; keyword; any other user-defined preference Analyzing the pattern in the data set, including receiving one or more of (ii) determining an estimated price from an incomplete historical price data set at any point in time for (i) above And (iii) the required price for goods or services that satisfy the parameters And calculating an estimate against, (iv) providing a price quote, is arranged server to perform is provided.

  In response to (ii), the server obtains past prices from the computer data store, (b) groups past prices by category, and (c) each group. Deriving statistical data for, (d) storing for each group a classifier containing the derived statistical data, and (e) identifying the group using the stored classifier corresponding to the requested price It may be arranged to perform.

  For (iii), the server uses statistical data obtained from stored classifiers corresponding to groups that the server has specified to calculate a set of estimates for the requested price over a specified date range. It may be arranged so as to perform.

  The server may be further arranged to perform the method of any method according to the first aspect of the invention.

  According to a third aspect of the invention, a computer program product embodied on a non-transitory storage medium, wherein the computer program product provides a price quote when executed on a computer. When the computer program product is executed on a computer, the computer program product receives (i) a request for parameters defining the goods or services together with a price for the goods or services, such as airfare, Includes action types such as airfare, hotel reservation, railway fare; date range; destination; origin; desired weather condition; star rating; keyword; one or more of any other user-defined preferences Receiving, (ii) at any point in time for (i) above, an incomplete past Determining an estimated price from the case data set by analyzing patterns in the data set; (iii) calculating an estimate for the required price for the goods or services that satisfy the parameters; and (iv) the price. A computer program product is provided that is arranged to provide an estimate.

  The computer program product may be arranged to perform a method belonging to any of the methods according to the first aspect of the invention.

  According to a fourth aspect of the present invention, there is provided a computer terminal connected to a server of any server according to the second aspect of the present invention, wherein the computer terminal is designated for a trip from a departure place to a destination. A computer terminal arranged to send a request for a price quote for a particular date range is further provided that is arranged to receive the price quote.

  According to a fifth aspect of the invention, an article or service, for example an air ticket, for which a fare estimate for the article or service is obtained using the method according to the first aspect of the invention or the second aspect of the invention. An article or service provided using a server according to the above aspect is provided.

  According to a sixth aspect of the present invention, a web-based price comparison web site, where an end user interacts with the web-based price comparison web site for a price for goods or services And a web-based price comparison web site that provides the request to estimate the price using the method of any method according to the first aspect of the invention. A web-based price comparison web site is provided.

  The foregoing and other aspects of the invention will now be described by way of example only with reference to the following figures.

FIG. 3 is an illustration of an embodiment of a system for estimating prices from other prices, with the purpose of using that price to increase the observed fare cash. FIG. 6 shows an example of how to use a “difference” engine to derive fare class availability from live updates. FIG. 4 shows an example of an expression suitable for use in naive Bayes classification. It is a figure which shows the Example by which a feature value is represented by the numbered area | region and represents frequency information with a price range (bucket). FIG. 6 illustrates an example of the entire training process. FIG. 7 is a diagram illustrating an embodiment in which each candidate quote placeholder can be priced using a naive Bayes classifier. FIG. 5 is a diagram illustrating an example of the entire process for inferring missing estimates. FIG. 7 illustrates an example of a connection between a server that provides a fare availability and price quote service, an input data source, and an output for partners and users. It is a figure which shows the Example of a user interface. It is a figure which shows the Example of a user interface. 1 is a diagram illustrating an example of a system for performing a search process. FIG. It is a figure which shows the Example of a user interface. It is a figure which shows the Example of a user interface output. It is a figure which shows the Example of a user interface output. FIG. 3 illustrates an example of an architecture for browsing a search. FIG. 2 illustrates an example of an overall system architecture for a system that includes providing fare availability and price quote services. It is a figure which shows a GDS structure Example. FIG. 6 is a diagram showing an example of a fare regarding a fare class record, a fare category, and a category data table. FIG. 6 illustrates an example of a process for providing a fare pricing option. It is a figure which shows the Example of the relationship between a price setting engine, AVS / NAVS, seamless (polling), cash DB, and an airline host. It is a figure which shows the system example containing an availability server. It is a figure which shows the graphic display example of the belief range with respect to the estimated price using a bar, and a bar covers the price range from the lowest price of the range which should find a price to the highest price. Bars are provided for the range of outbound dates and the range of return dates.

  One implementation of the present invention provides “model-based pricing”. This is described in Section A below.

A. Model-based pricing Flight price estimation using model-based data interpolation is provided.

Background / Motivation Ticket prices for itineraries are complex to calculate. The result of the calculation of possible routings, authorized fares and the availability of these fares. These calculations are typically performed in a global distribution system (GDS), and executing queries against such systems can be slow and expensive.

Airlines that file fares in the GDS typically use a pricing model called “scheduled”. The sum of such important things should be noted about the model, if the round-trip travel the day O ₁ to return to the starting day R _1, the price of round-trip ticket P (O _{1, R 1),} the two one-way ticket That is, it is not equal to P (O ₁ ) + P (R ₁ ). An alternative pricing model (referred to as “low budget” pricing) is a model where P (O ₁ , R ₁ ) = P (O ₁ ) + P (R ₁ ).

  Two important consequences of the scheduled pricing model are the restrictions imposed on observed price reuse and the space required to store them.

In the low budget model, two “round trip” queries are made, where one is for the date O ₁ R ₁ and the other is for the date O ₂ R ₂ , the resulting price P (O ₁ ), P (R ₁ ), P (O ₂ ), and P (R ₂ ), up to four round-trip fares (ie, P (O ₁ , R ₁ ), P (O ₁ , R ₂₎ ), P (O ₂ , R ₁ ), P (O ₂ , R ₂ )) and four fares P (O ₁ ), P (R ₁ ), P ( O ₂ ), and P (R ₂ )). If the query is run against a “scheduled” model, there will be only two fares, namely P (O ₁ , R ₁ ) and P (O ₂ , R ₂ ). Thus, in the low budget model, eight queries can be answered by two queries, whereas in the scheduled model, two queries can be answered by two queries.

Assuming that the flight price is available only on the current day and the next 364 days, finding all possible one-way and round-trip fares on the route is possible by making 365 round-trip inquiries. Become. For a scheduled model, 365 ^* (365 + 1) / 2 = 66,795 round trip queries as well as 365 one way queries (in each direction) would be required. Thus, to build a complete list of prices for a route, the schedule model requires about 185 times more queries than the low budget model.

  Having the results of all these queries will inevitably increase the storage space required to store the fare approximately 185 times.

The foregoing is just an example for a route between two defined airports. In fact, there are about 5,000 private airports in the world. Therefore, the number of possible trips between airports is about (5,000) ² . Considering the number of possible round-trip queries of the number 365 ^* (365 + 1) / 2 = 66,795 mentioned above, this gives (5,000) ^{2 *} 365 ^* as the total number of possible queries between all airports per year (365 + 1) / 2 is obtained, which is about 10 ¹² . The implied amount of data storage is actually very large.

  Combining scheduled model paths with GDS costs greatly increases the number of queries needed to form a “big picture”, and building such a cache is not practical (ie, inconvenient) Is). And the result is incomplete cash, and there are “holes” for which no fares have been observed.

  However, the price (returned by GDS) is the result of applying fare rules and availability. As a result, there are patterns and consistency in the observed fares. This embodiment is capable of interpolating the fare missing from the cache based on using these patterns, using a model that reflects the fare rule that generated the price at the first place.

A system for estimating prices from other prices A system for estimating prices from other prices, using that price to be observed (eg, requesting a price for a live bookable fare) With the purpose of increasing the fare cash (observed by). A model embodiment is shown in FIG.

  The estimator predicts other estimates using statistical data derived from the estimates. A very simple estimator might calculate that all missing fares are equal to the average of all observed fares. The level of prediction can be improved by making the model more sophisticated to take into account the type of rules used to generate the fares in the first place.

  For example, a common “rule” governing whether a fare is valid for an itinerary is whether the itinerary involves staying on a Saturday night. Typically, the meaning of such a rule is that a date with not spending a Saturday night at the destination is much more expensive. The rules are designed to penalize business travelers who normally want to return during a work week. In this way, a better model containing two average values can be calculated. That is, the average value of all observed fares with a Saturday night stay and the average value of all observed fares without a Saturday night stay. The algorithm then predicts the price for a date pair that has not been observed so far, by first calculating whether there was a Saturday night between date pairs, and then returning the corresponding average Do. As a result, the accuracy of estimation increases.

  Similar improvements can be made to the model. For example, the price of a trip varies according to the day of the week (Monday is typically more expensive than Wednesday). Thus, calculating and using an average that also takes into account the day of the week will still potentially improve model accuracy.

  Accordingly, providing a statistical model for predicting fares may be performed using, for example, statistical data and rules that may be derived by a human. Alternative methods of forming a statistical model are possible. In this approach, a machine learning approach called “naive Bayes classifier” is used to form a fare probability model and that model is used to predict unobserved fares. The classifier is trained using a set of observed prices and corresponding “features”. Features could include the departure date of the week, number of days stayed, stay on Saturday, airline, etc. for the query that generated the fare. The classifier then predicts the price of the unobserved fare by being given a set of features and giving the price most likely to have those features.

  The statistical model for predicting the price can then be used as a reference to predict fares that have not been observed so far, or to predict the observed fares (if it is desired to measure the accuracy of the model).

Key features and their computation Features that may be used to train the classifier include:
・ Saturday stay ・ Number of days of stay ・ Day of the week ・ Time to travel ・ Airline ・ Website ・ Route ・ Date-day and year-month level ・ Month

  These features analyze all the fares listed in FROP (Fare Rule Output Product, from ATPCO (Airline Customs Publisher)) and determine which rule applies to the maximum number of fares It is derived by doing. For example, 54% of the fares have a minimum stay rule, and 31% of all fares require a Saturday night to be spent at the destination as a minimum stay rule. Thus, features can be useful features for classification based on price. Deriving a more accurate estimate of the usefulness of a feature may be done by training multiple models with different features and comparing the prediction accuracy of different models.

Estimate profits, missing flight price data, and create superior forecast models using data generation features

Potential market and / or application area models may be used in the flight industry to estimate previously unobserved prices, display prices to the user, or inquire to return prices It may be determined whether there is (for example, there is a possibility of a good value).

  Although this model has the disadvantage of obtaining prices (eg, inquiry costs) as another field, it may be used in fields where price generation is performed by structured means (eg, pricing rules). This includes travel products such as hotels, trains and rental cars, and other products with structured pricing such as insurance.

Other Applications Consumer price comparison web site companies attempt to price competitively by comparing prices with competitor prices.

Model-based pricing-Estimating missing estimates Estimating missing estimates may be done in two stages (training and estimation).
・ Training ○ Building a statistical model from past prices ・ Estimating ○ Identifying missing estimation candidates ○ Price estimation candidates based on statistical models

  In an embodiment, the statistical model is a set of naive Bayes classifiers that are trained regularly (eg, daily) using past prices. The missing estimate is inferred based on the route operator data and priced using a classifier.

Examples of statistical models Naive Bayes classification is a dependency between the probability that an object described by a set of features belongs to a class and the probability that a set of features from that class is characterized. Based on above. See equations (1) and (2) in FIG.

In FIG. 3, equations (1) and (2) can be simplified to equation (3). In the formulas (1) to (3),
C is a class that can classify the input object (for example, in this case, a price range, for example, 75 to 125 EUR).
_Fi is a feature value (for example, in this case Friday, Air France Airlines website).

  The classification can be expressed using equation (4) in FIG.

  For a set of feature values where it is desired to find a class (price range), feature values occur most frequently, so the probability that the class will characterize the input set is maximized.

Training The intent of classifier training is to calculate how often feature values occur for the observed price range. For example, in the process shown in FIG. 4, feature values are represented by numbered areas, and price ranges (buckets) represent frequency information (for example, feature value quantities range from 75 to 125 EUR, 126 to 175 EUR, etc.). .

The following elements are entered for the training process example:
・ List of paths ・ Classifier category classification method ・ Past estimates ・ Supported feature pairs with weights

  The list of paths sets the boundaries for training. A category defines a range of a classifier, and the category is expressed by a set of feature values. The experimentally selected categories consist of city-level routes and airlines (eg, London to Paris, Air France), but other options are possible (eg, city-level routes without airlines). There is always one classifier per category. The classifier is trained using only past estimates that match that category (eg, only estimates for the London-Paris route (operated by Air France)).

  In the first step of the training process, a certain number of the latest historical estimates with some maximum age is taken from the database. Evaluate the category against any estimate. Group estimates by category. Each group becomes a training set for one classifier.

  The training set can be expanded with estimates for the reverse path equivalent. Example: In the case of historical estimates on the route London Stansted to Rome Ciampino (Air France), the reverse route equivalent is Rome Ciampino to London Stansted (airline and other details such as price The same). An estimate for the reverse path is included in the training scaled by weights that adjust its importance.

Next, a classifier with the following statistical information extracted from the estimate is constructed for each group.
・ Weighted number of estimates including each different feature value ・ Weighted number of estimates including each different feature value per price range of a certain granularity ・ Many observed price ranges, estimated price ranges, and all estimates

  The weights applied to many estimates can be varied between features to emphasize that some of them are more affected (for example, it may be more important to include weekdays in the trip than the length of stay) is there).

  The process ends with one classifier for each training set that includes statistical information (eg, one classifier for the London-Paris route (operated by Air France)). Store model for future use. The classifier can be retrained, for example, periodically or when many past estimates have changed.

  A classifier may be formed only if a certain number of estimates matching that category are available.

  An example of the entire training process is shown in FIG.

In addition to the parameters listed above, the training process can be parameterized by:
-Price range size per route (eg 1EUR, 5EUR, 10EUR)
・ Weighting function applied to reverse path ・ Weighting function applied based on estimated age

  Estimate missing quotes

  In an embodiment, the user searches for a one-way or round-trip flight for a certain route and date range. The system responds to the query with a set of cheapest estimates obtained from the database. At most one for each date (or for date-to-round flights). The response may not include an estimate for a date (date pair). Guessing the missing estimates can be done using a trained naive Bayes classifier.

The following elements may be entered into the guess process:
・ List of dates (or date vs.) with route / missing cheapest estimate information

  Since the classifier can only be used to price existing quotes, the candidate quotes must be evaluated first. Candidates are any applicable airport pair, date (date pair), airline, and agency combination available for the query path. The list is generated based on the process input and the route operator database.

  Next, each candidate quote placeholder is priced using a naive Bayes classifier. For example, see FIG.

Price estimation may be performed for each candidate estimate in the following steps.
• Extract category feature values from estimates.
・ From the database, extract the classifier trained for the extracted category. ・ Extract all feature values from the estimation candidates. ・ Categorize the candidate estimation by the following: ○ For each price range stored in the classifier Calculate the Bayesian posterior probability against ○ Select the price range class with the highest Bayesian posterior probability ・ Put the price class into the candidate estimate

  If there is no classifier for a candidate estimate, no price estimation is performed and the date (date vs.) slot remains empty.

  The process can result in multiple priced quotation candidates per date (date vs. date). For this reason, the next step is to select a single quote for each date (date pair). Selection can be made based on price. For example, select the cheapest estimate per date (vs. date).

  The end result of the process is at most one estimate per input date (date vs.). The estimated estimate can be included in the response returned to the user. An example of the entire process for inferring missing estimates is shown in FIG.

In addition to the parameters already listed, the guessing process can be parameterized by:
• Minimum Bayesian posterior probability necessary to allow classification results • Maximum number of path operators involved in candidate generation • Random variation added to Bayesian posterior probability to avoid ties

Estimating method that gives certainty to price As an estimating method, one that gives certainty to price may be provided. For example, 99% are convinced that it is within $ 5 and $ 1000. Thus, the estimation method may provide:
• Calculation and return of these beliefs.
Use these beliefs to determine whether to display prices to the user or indicate to the user a range of possible prices (eg, error bars).

  Many ways to estimate prices can return a measure of confidence in the price. For example, in the case of a naive Bayesian classifier, the selected class is the class with the highest probability, but that probability can be used as an indicator of confidence in the estimate. The closer the probability is to 1, the higher the certainty that the predicted price is more accurate.

The resulting belief may be used in many ways. If the confidence in the estimate is very low, the following decision may be made.
O Don't give an estimate because it's not useful to the user o Trigger the live retrieval of the price so that the exact value can be shown to the user. Confidence can be shown to the user.

Confidence can be shown to the user. This can be shown in many ways.
O A bar indicating the range of possible values for the price (see, for example, FIG. 22).
Reduce the visibility / intensity of the displayed price (eg, using coloring, shading, or dotted lines).
The confidence range for the estimated price ranges from the lowest price to the highest price range that leads to the confidence that the result of the estimation method will find the price. The confidence range for the estimated price may be shown graphically using a bar. A bar may be displayed for each date in the date range. Displaying the bar for each date in the date range may be performed for the forward and return paths. Bars are color-coded, for example, bars correspond to recent prices, for booking functions, for price ranges derived from multiple tickets sharing a departure date, and similar itineraries The price may be stable with respect to the price range estimated from One bar may be provided for a certain date. A plurality of bars may be provided for a certain date. An embodiment is shown in FIG.

B. Derived Availability Engine Background / Motivation Motivation is very similar to that in model-based pricing embodiments.

  The meaning of the “scheduled” pricing model is that it is not possible to form a round trip quote using two single quotes, so a query must be performed for every possible date pair, This is prohibited because of the large number of pairs, due to the cost of obtaining this data and the space required to store the data. However, the round trip price is the result of combining the available fares available for the outbound and return legs, along with some rules. Thus, if the availability and rules can be known, it is possible to break down round trip queries into “reusable” legs and use them to construct new “round trip” prices. Is possible in a similar (but different) way to that used for airlines with

Derived Availability Engine Example The World Distribution System uses schedules from OAG (Official Airline Guide: see www.oag.com), fares from ATPCO, and fares classes from airlines Gather possibilities and store this information in its own system. This stored information is periodically updated from those sources. When an inquiry is made to GDS for an estimate for a route and date, GDS performs the following steps.
1. Determine a valid itinerary from the schedule.
2. Calculate a valid fare for the itinerary (some GDS do not have a pricing engine and are provided by a third party engine, eg SITA (see eg http://www.sita.aero/) ) To ensure that the correct price applies for each travel itinerary).
3. Find the availability of these fares.
4). Add accurate taxes and surcharges.

  Step 1 may be implemented in the route service. Specifically, step 1 may be realized using a timetable and a rule about which can be sold. When there is a user inquiry, an effective itinerary that can purchase a ticket can be specified.

  The airline sends the fare to ATPCO, which provides an application for these data. However, it would be a great effort to convert these outstanding fares and rules into a system for determining the exact fares for a query. Fortunately, ATPCO supplies merged fare and rule data in a data supply called FROP (Fare Rule Output Product). The FROP data is transmitted in a fixed length recording file containing rule conditions and fare information compiled for the main categories. Therefore, step 2 can be performed using FROP data. It can also be realized by combining fare rules and prices from airlines using a pricing engine.

  ATPCO also provides a supply of itinerary and service charges taxes. As a tax and surcharge engine, take the data in these applications and use them to calculate the total price that the customer will pay for a fare, thus building an engine that emulates step 4 be able to.

  The only remaining item for low cost calculation of GDS fares is availability-step 3. It is possible to derive the availability by comparing the result of the live update (final result of steps 1 to 4) with the calculations of steps 1, 2 and 4 outlined above, ie route service, FROP. All possible fares can be calculated by combining the fares obtained from and the tax and surcharge engine, and using the observed fares obtained from live updates, It can be inferred which one was used and thus which of the calculated fares were available. The diagram in FIG. 2 shows how the availability of fare classes can be derived from live updates using such a “floor” engine.

  If the available fare is known for both the outbound and return leg of a round trip flight, the round trip price can be calculated even if a specific date pair in the round trip has not been observed. This method of modeling estimates is known as connectable single leg (SLC). SLC is similar to the “low budget” pricing model. However, in the case of a low budget, the return trip and return trip legs are also effective as one-way tickets. For low budgets, constructing a return trip is simply a matter of adding inbound and outbound leg prices (since the SLC structure needs to apply availability after finding a valid fare for a date pair, Slightly more complicated).

  If the price of the fare has increased but the availability remains the same, it is possible to calculate a new price using the derived availability engine, even though it has not been observed.

  There are additional benefits to doing this provision of derived availability compared to using GDS. Although a pure calculation in GDS does not guarantee that the fare can be “found” on the airline or OTA website, the observed fare is by definition found. Pure calculations also do not take into account surcharges, margins, discounts, and personal fares (which may be incorporated into the observed fares).

  Note that the fare price does not necessarily change, and there are many different fares for a specific flight, and the availability of each fare may change. Revenue management software may continuously change which fares are available, so a cheaper ticket can disappear because revenue management cancels that ticket and increases revenue .

Derive availability, not need to contact GDS or airline reservation system to find profit / availability.
Calculate new prices based on estimated availability rather than true availability.
• Estimate new prices based on availability and fare rules without accompanying observed cases of new prices.

Other potential applications To provide airlines with pricing for their own products. This is because airlines are still doing this pricing from GDS at a high cost.

  The company is interested in finding flight data prices at a fairly low cost.

Architectural Overview Embodiment A server (which may be real or virtual) is provided that provides fare availability and pricing services. Related websites, smartphone apps, business-to-business services, connectivity with traditional airlines (eg British Airways, Qantas, KLM, etc.), low budget airlines (eg Ryanair, EasyJet, Germanwings, etc.) And a travel reservation web site (eg, lastminute.com, opodo, Thomson, etc.).

  Travel booking websites and airlines may provide input data to servers that provide fare availability and price quote services. The server may provide the output to a public application programming interface (API) such as a partner company. The server may provide the output to the user for the user to provide the output to the search web site, to the search mobile website, and to the app on the smartphone. An example of a connection configuration between a server and an input data source, output for partners and users is shown in FIG.

  In a search at the server, the user may specify a departure location (eg, can be an individual airport, a city containing multiple airports, or a country containing multiple airports). In a search at the server, the user may specify a destination (eg, an individual airport, a city containing multiple airports, or a country containing multiple airports). For searches on the server, the user specifies a departure date or a range of departure dates (for example, a date range, or calendar month, or an unspecified date (for example, can include a date range such as any date in the following year)) You may do it. For searches on the server, the user specifies a return date or a return date range (for example, a date range, or calendar month, or an unspecified date (for example, can include a date range such as any date in the following year)) You may do it. In a search on the server, the user may specify a one-way ticket or a round-trip ticket. The user may specify the number of passengers in a specific age range. The user may specify a ticket class (eg, economy, premium economy, business class or first class, etc.). A parallel search for hotels may be performed. A parallel search for a rental car may be performed. You may also show your preference for direct flights. A user interface embodiment is shown in FIG. A further user interface embodiment is shown in FIG. This is an example of a flexible search. Because departures indicate that any airport in London (UK) can be used, and destinations indicate that any airport in Australia is allowed.

  The output search result may include a list of flights that satisfy the search criteria. A graphical indicator (eg, a slider bar) may be provided to limit the departure time range for outbound flights. A graphical indicator (eg, slider bar) may be provided to limit the departure time range of the round trip flight. Selectable tabs may be provided so that flights are arranged in order of increasing cost. Selectable tabs may be provided so that flights are arranged in order of increasing travel time. Selectable tabs may be provided to arrange the flights in the order in which the airline names are arranged in alphabetical order. A user interface embodiment is shown in FIG.

  In the search example by the user, if the flight between the departure point and the destination (which may be a one-way flight or a round-trip flight) corresponds to the flight date to the foreign country, Required for round trip flight dates. In the search process embodiment, in a first step, a partner search web site that exhibits output is selected. In the second step, the associated prices already cached are identified. In the third step, relevant prices that are not available are identified and updated. In the fourth step, the new data is cleaned and stored (eg, taxes and fees may be removed and the actual flight operator may be identified). In the fifth step, the price was returned to the user. An example of a system for performing the search process is shown in FIG.

  In a flexible search, the output data provides a list of destination airports that meet the flexible search criteria (for example, the destination is any airport in Australia and departs from any airport in London, UK). good. The list may be listed in ascending price order. An example of user interface output is shown in FIG. Once a particular airport has been selected, the minimum price for the outbound trip leg may be shown as a function of the outbound departure date, for example for a selectable calendar month, and the minimum price for the inbound trip leg may be displayed As a function of day, for example, for a selectable calendar month, it may be shown for a selected departure and destination city or airport. An example of user interface output is shown in FIG.

  A browsing search may be provided. A browsing service may be provided to users of smartphone apps or users of search web sites. The browsed price data may be accessed by a browsing service. The viewed price data may include data scraped from the website content. The browsing service may be connected to a public application programming interface (API). The browsing service may not be connected to the airline server, or may not be connected to the server of another travel information service provider. An example of an architecture for browsing search is shown in FIG.

  An example of an overall system architecture for a system that includes providing fare availability and price quote services is shown in FIG.

  The following sections provide background information on how to calculate airfare and the role of GDS.

C. Introduction to the Global Distribution System The Global Distribution System (GDS) may be defined as a system that includes availability, price, and related services for airlines, through which reservations and tickets can be issued. . With GDS, these functions are made available to subscribed travel agencies, booking engines, airlines, and other applicants. A GDS configuration embodiment is shown in FIG. A more general term is “distribution system”. This may include making reservations for modes of transport other than air transport (eg, trains, etc.).

Route Search Historically, route search was added by GDS as the airline network grew. When the number of potential possible connections grows exponentially (the SOF-NYC has many connection points. For example, SOF-IST-NYC, SOF-FRA-NYC, SOF-MOW-CPH- NYC, etc.), GDS added this “connected” building layer to reduce the number of different priceable routes. Therefore, it is not necessary to process many fares and rules. They did it with a very dull method-static route and connection table. This method tells them essentially a SOF-NYC routing city to fly between zones (defined by ATPCO-US-EUR, EUR-APAC for simplicity of discussion). However, potentially including connections in the wrong direction (eg IST), but reject many cities as possible connections. For example, flight SOF-PRG-CPH-NYC may be considered too expensive and potentially expensive and is therefore dropped during the route search phase.

Connection Setup After setting up the routing, the connection setup begins to look up the airline, number of flights, and connection limits (included in the Minimum Connection Time (MCT) table). MCT (distributed by OAG) is a complex document that specifies general rules on how flights / airlines can connect in terms of time. For example, if a long distance US-EU connects to the EU-EU in the LHR, a 120 minute MCT is applied. General rules are easy to interpret but important. If someone tries to book outside the range of MCT parameters, manual invalidation is required.

  Of interest is the MCT exception (about 200,000). This may take up to 55 minutes until DL flight # 003 connects to AF flight # 004. However, when DL flight # 003 arrives at the CDG gate 19 and the connecting flight AF009 departs from gate 49, 65 minutes is applied as the MCT, and so on. About 70,000 of all exceptions are to Air France (AF) and Paris-Charles de Gaulle Airport (CDG). That's why many passengers think it is the worst to fly.

Fares An example of fares is a pricing contract. This specifies the price (basic fare) for services provided by the airline. Prices are defined by service class, class type (Y, Q, J, M, H, etc.) and type of trip (one way, round trip, multi-city, round the world (RTW)) .

  In an embodiment, each fare is accompanied by the number of rules and restrictions that apply to it, and the fare is not a simple Excel row (spreadsheet or CSV file), but a 20-30 page long contract. Think of it as Rather than thinking of the fare as a row in a large Excel spreadsheet and the columns indicating what features the fare has, instead, the fare is itself accompanied by many parameters and restrictions It should be considered a complicated matter, and is rather close to a contract. The fare is when, how and when the fare-passenger type can be used, when the fare is applicable (NVB, NVA-not valid before, not valid after), and many others Stipulates restrictions. The fare can be a public fare (accessible to any reseller) or a personal fare (accessible to a designated reseller / seller).

  Fares are given by the airline. They are defined at a general level for the market and are limited by travel and ticketing date. Naturally, fares represent price points.

Fare Upload and Distribution Upload and Distribution ATPCO is an industry association that handles fare and fare distribution in the industry. There are a few airlines that only deliver to SITA (Bulgarian Airlines is an example).

Scheduling Up to now, ATPCO has released fares on a specific schedule, but at the request of the airline, the organization has begun to release fares every hour. This allows airlines to ensure a quick response to market changes that are not only suitable for availability control.

  SITA will release the fare four times a day on weekdays and three times on weekends. There are also fares that persist within the GDS (GDS-specific fares or negotiations). This fare is completely under GDS control and does not follow the general delivery schedule.

Fare rules and restrictions and categories Provide fare data, fare class data, and fare restriction data.

How pricing works (very high level)?
Rule calculation is performed by the pricing engine in a very strict order of rule records, tables, and sub-tables. If specific rule records are not found, there are general rules that prescribe them all.

Restricted data-The fare class tells the engine that there are rules for a particular fare.
Category control tells the engine what type of restrictions exist.
-The category data table is last when the engine reads to get relevant information.

  For example, a fare class tells the engine that a rule is attached. The rule is category 15 (sales restrictions), which, after examining the category data table, derives additional information about it (what type, what are the rules and restrictions).

  An example of fare for a fare class record, fare category, and category data table is shown in FIG.

Fare display vs. fare estimate and process Many industry experts use the terms fare display and fare estimate interchangeably, which is unacceptable to the skilled worker. This will be described.

Fare display A comprehensive integrated concept of market fares and related rules. This allows the agency to:
-Compare fare rules (NB! Not price)
-Telling passengers the contract period (fare rules)

Fare Estimate Final price after applying the contract period. This captures most of what we want to give passengers.
-All taxes-All surcharges-All rules associated with it-Final price

  An example of a process for providing a fare pricing option is shown in FIG.

Availability Knowledge of availability is very important in the industry. Depending on availability, the airline can control its margins (seats) and add another way to manage how profitable the route is by opening and closing the availability bucket Made. This is managed by a revenue management system (RM).

  Some RMs have a high ability to manage complex rules, while complexity is high because most of the availability rules are formed by internal departments of statisticians and mathematicians.

  There are three main ways to receive availability when it is desired to check whether a seat / class service is available.

i) NAVS / AVS
NAVS = availability by simple numerical value per seat class. Example: Y9H3M0L8. . .
AVS = Availability Status-This is a flight open / closed, open to a waiting list, etc.

ii) The polling CXR (airline) can use NAVS / AVS as a display if a better connection drops. These CXRs use revenue management rules and need to be polled to get actual availability and all availability information.

  Some airlines and / or flights require polling, and the order for doing so is MR (Market Restriction). If the tag is MR when the GDS requests AVA, the GDS needs to poll for the correct availability. Unfortunately, NAVS / AVS does not show these tags, so if GDS uses NAVS / AVS for tagged flights, it may display the wrong availability.

iii) Cache DB
In order to reduce polling and NAVS / AVS, the pricing engine has evolved a complex cache DB. Although they are not accurate, they achieve some of the goal of reducing calls when storing information.

  An example of the relationship between the pricing engine, AVS / NAVS, seamless (polling), cache DB, and airline host is shown in FIG.

Availability Server Embodiment The caching mechanism may be supplied by offline traffic, but is used by the online channel. The availability server may be able to update the cached data based on rules (bidirectional polling) and also warn of minor discrepancies with respect to reservations. A system embodiment including an availability server is shown in FIG.

NOTE Performing the steps described above can be done using standard well-known programming techniques. The novelty of the embodiments described above involves achieving the described results using the described steps, rather than being in a specific programming technique. Software programming code that embodies or forms part of the present invention is typically stored in permanent non-transitory storage. In a client / server environment, such software programming code may be stored using a storage device associated with the server. The software programming code may be embodied on any of a variety of known media (eg, diskette, hard drive, or CD-ROM) for use with a data processing system. The code may be distributed on such media, or distributed to a user from one computer system's memory or storage device to another computer system over a certain type of network. You may make it usable for the user of such other systems. Techniques and methods for implementing software program code on physical media and / or distributing software code over a network are well known and will not be described further herein.

  It will be appreciated that performing each element of the description and combinations of elements in the description, a general purpose and / or dedicated hardware-based system that performs the specified functions or steps, or general purpose and / or dedicated hardware This can be done by a combination of hardware and computer instructions.

  These program instructions may be provided to the processor to form a device such that the instructions executed on the processor form a means for performing the functions specified in the description. Computer program instructions are executed by the processor to cause the processor to perform a series of operational steps to form a computer-implemented process that performs the functions specified in the description when the instructions are executed on the processor A step may be given for this purpose. Accordingly, FIGS. 1-17 and 22 support combinations of means for performing the specified functions, combinations of steps for performing the specified functions, and program instruction means for performing the specified functions. .

  Of course, the above arrangement merely illustrates application to the principles of the present invention. Many modifications and alternative arrangements can be devised without departing from the spirit and scope of the present invention. While the invention has been illustrated in the drawings, with individuality and details, and fully described above in connection with what is presently considered to be the most realistic and preferred embodiment of the invention, it will be apparent to those skilled in the art. Obviously, numerous modifications can be made without departing from the principles and concepts of the present invention as described herein.

Concepts Multiple concepts (hereinafter referred to as concepts “AH”) are included in this disclosure. The following may be useful in defining these concepts. You may combine the aspect of a concept.

A. A method for providing a price quote, comprising: (i) receiving, by a computer server, a request for a parameter defining the goods or services along with a price for the goods or services, eg, airfare; (ii) ) At any point in time for step (i) above, determining one or more processors to determine an estimated price from an incomplete past price data set by analyzing a pattern in the data set (Iii) configuring one or more processors to calculate an estimate for a requested price for goods or services that satisfy the parameters; and (iv) configuring the price estimate to an end user computing data. Sending to a device such as a personal computer, smartphone or tablet.

The foregoing may further include any of the following alone or in combination.
-In this method, the parameters that define the goods or services include action types, such as airfare, hotel reservation, railway fare; date range; destination; origin; desired weather condition; star rating; keyword; One or more of the user-defined preferences.
• Estimate price is determined by inferring, deriving or predicting the estimated price.
Step (ii) includes (a) obtaining past price estimates from a computer data store; (b) grouping past price estimates by category; and (c) statistical data for each group. (D) storing a classifier including the derived statistical data for each group on the computer; and (e) using the stored classifier, a group corresponding to the required price is determined. Identifying.
Step (iii) includes calculating a set of quotes for the requested price over a specified date range using statistical data obtained from the stored classifier corresponding to the specified group. It is.
Step (ii) includes analyzing patterns in the data set using rules.
Step (ii) includes a naive Bayes classifier machine learning approach that generates a probabilistic model of price, and uses that model to predict unobserved prices.
Train the classifier using observed prices and their corresponding feature sets.
Features are related to the request and include one or more of the departure date of the week, the number of stays, stay on Saturday, airline, travel time, route, month.
The classifier then predicts the price of the unobserved price by being given a set of features and giving the price most likely to have those features.
The feature may be derived by training a plurality of models using different features and comparing the prediction accuracy of different models.
Step (ii) includes building a statistical model from past prices, identifying missing quote candidates, and pricing quote candidates based on the statistical model.
• Estimating the price for each candidate estimate, extracting a category feature value from the estimate, extracting a trained classifier from the database for the extracted category, Extract feature values, classify candidate estimates, calculate Bayesian posterior probabilities for each price range stored in the classifier, and select the price range class with the highest Bayesian posterior probability And the step of: attaching a price class to the candidate quote.
Input to the statistical model includes a list of paths, a classifier category classification scheme, a past estimate, and a set of supported features with weights.
-Filter past estimates by age.
• Inputs to the statistical model include backward path equivalents.
The method includes the step of including the cached fare price in the set of price estimates.
The set of price estimates is constructed without inquiring the distribution system after step (i).
• Prices are for trips that are one-way trips.
• Prices are for trips that are round trips.
・ The price includes the air fare price.
・ Price includes railway fare price.
・ Price includes rental car price.
・ The price includes the hotel price.
The request includes a flexible search request.
The final result of the method is at most one estimate per input date or input date pair.
The method includes (A) at any point in time for step (i), determining the confidence range of the estimated price from an incomplete past price data set by analyzing the pattern in that data set. Configuring the one or more processors in such a manner and (B) configuring the one or more processors to calculate an estimated price confidence range for the required price for the goods or services that satisfy the parameters. It is.
The method further includes the step of (C) providing a fare price estimate along with a confidence range to an end user computing device, such as a personal computer, smartphone, or tablet.
The method further includes the step of determining using the confidence range whether to display the price to the user or provide the user with a likely price range.
-Display possible price ranges as error bars.
• The date range includes only one departure date.
• The date range includes only one return date.
The method includes a method that provides a set of best round-trip fare price estimates for a specified departure date range and a specified return date range, and step (i) includes for a trip from the origin to the destination. Receiving a request for a round-trip fare price for a specified departure date range and a specified return date range, and step (iv) includes determining a best round-trip fare price estimate for the specified departure date range and the specified return date range; Providing a pair is included.
• The best round-trip price quotes are provided in bar chart format.
Of the minimum Bayesian posterior probabilities necessary for the estimation process to accept classification results, the maximum number of path operators involved in candidate generation, or random variations added to the Bayesian posterior probabilities to avoid ties Parameterized by one or more.
The method is performed on a server that provides a fare availability and price quote service.

  A server configured to provide a price quote, wherein: (i) receiving a request for parameters defining the goods or services together with prices for the goods or services, such as airfare; Receiving, including one or more of any other user-defined preferences, such as airfare, hotel reservation, rail fare; date range; destination; origin; desired weather conditions; star rating; (Ii) determining an estimated price from an incomplete past price data set at any point in time for (i) above by analyzing a pattern in the data set; (iii) parameters Calculate an estimate for the required price for goods or services that meet If, (iv) providing a price quote, it is arranged server to perform is provided.

The server may further include any of the following, alone or in combination.
(Ii) the server obtains past prices from the computer data store, groups past prices by category, derives statistical data for each group, derived statistics for each group It is arranged to store a classifier including data and to identify a group using the stored classifier corresponding to the required price.
The server is arranged to use the statistical data obtained from the stored classifier corresponding to the identified group to calculate the set of quotes for the requested price over the specified date range.
The server is further arranged to execute a method belonging to any of the aspects of concept A;

  A computer program product embodied on a non-transitory storage medium, wherein the computer program product is arranged to provide a price quote when executed on a computer, the computer program product comprising: When executed on a computer, (i) receiving a request for parameters that define the goods or services along with prices for the goods or services, such as airfare, where the parameters include an action type, such as airfare, hotel reservation Railway fare; date range; destination; origin; desired weather conditions; star rating; keyword; including one or more of any other user-defined preferences, and (ii) as described above ( Estimated from incomplete historical price data sets at any point in time for i) Determining the case by analyzing patterns in the data set; (iii) calculating an estimate for the requested price for the goods or services that satisfy the parameters; and (iv) providing a price estimate. And a server arranged to perform the above.

  The computer program product may be arranged to perform the method of any aspect of concept A.

  A computer terminal connected to the server of any server of concept A, the computer terminal being arranged to send a request for a price quote for a specified date range for a trip from the origin to the destination, the computer The terminal is further provided with a computer terminal arranged to receive the price quote.

  An article or service, such as an air ticket, where a price estimate for the article or service is provided using a method defined in any aspect of Concept A or a server defined in any aspect of Concept A Goods or services are provided.

  A web-based price comparison web site where end users can interact with the web-based price comparison web site by providing a price request for goods or services; A web-based price comparison web site is provided that provides the request to a server that estimates the price using a method belonging to any aspect of concept A.

B. How to guess which fare class is available for a trip How to guess which fare class is available for a trip from a departure point to a destination on a specific date. The method includes the steps of (i) defining a fare price requirement for a trip from a departure point to a destination on a specific date; and (ii) calculating an estimated fare price from an incomplete historical price data set. Inferring, estimating or predicting by analyzing patterns in the data set; (iii) using the result of step (ii) to calculate an estimate for the requested fare price; iv) sending the request to a delivery system for the fare price for the trip from the origin to the destination on a specific date; and (v) a delivery system for the trip from the origin to the destination on a particular date. Receiving the fare price of (vi) and the requested fare price obtained from (vi) step (iii) Comparing the calculated estimate to the fare price of the distribution system received in step (v) to infer which fare class is available for the trip from the origin to the destination on a specific date And (vii) outputting an estimated fare class availability for a trip from the departure point to the destination on a specific date.

The foregoing may further include any of the following alone or in combination.
In the method, step (ii) includes obtaining past fare estimates from a computer data store, grouping past fare estimates by category, deriving statistical data for each group; On the computer, for each group, storing a classifier that includes the derived statistical data and using the stored classifier to identify the group that the requested fare price corresponds to .
In the method, step (iii) includes calculating statistical data from the stored classifier corresponding to the specified group to calculate a set of estimates for the requested fare price over a specified date range. To do with.
・ The distribution system is a global distribution system.
• Estimated fare class prices are attached to each estimated fare class availability.
Includes a step to send output to the server.
-The server is an airline server.
Step (ii) includes analyzing patterns in the data set using rules.
Step (ii) includes a naive Bayes classifier machine learning approach that generates a fare probability model, and uses that model to predict unobserved prices.
Train the classifier using observed prices and their corresponding feature sets.
Features are related to the request and include one or more of the departure date of the week, the number of stays, stay on Saturday, airline, travel time, route, month.
The classifier then predicts the price of the unobserved fare by being given a set of features and giving the price most likely to have those features.
The feature may be derived by training a plurality of models using different features and comparing the prediction accuracy of different models.
Step (ii) includes building a statistical model from past prices, identifying missing quote candidates, and pricing quote candidates based on the statistical model.
• Estimating the price for each candidate estimate, extracting a category feature value from the estimate, extracting a trained classifier from the database for the extracted category, Extract feature values, classify candidate estimates, calculate Bayesian posterior probabilities for each price range stored in the classifier, and select the price range class with the highest Bayesian posterior probability And the step of: attaching a price class to the candidate quote.
Input to the statistical model includes a list of paths, a classifier category classification scheme, a past estimate, and a set of supported features with weights.
-Filter past estimates by age.
• Inputs to the statistical model include backward path equivalents.
The method includes a step of including the cached fare price in the fare price estimate set.
・ Travel is a one-way trip.
・ Travel is a round trip.
-Fares are air fares.
The request includes a flexible search request.
Of the minimum Bayesian posterior probabilities necessary for the estimation process to accept classification results, the maximum number of path operators involved in candidate generation, or random variations added to the Bayesian posterior probabilities to avoid ties Parameterized by one or more.
The method is performed on a server that provides a fare availability and price quote service.

Further, a server configured to infer which fare classes are available for a trip from a departure point to a destination on a specific date, the server comprising: (i) a departure on a specific date Data that specify requirements for fare prices for travel from place to destination, and (ii) infer, estimate, or predict estimated fare prices from incomplete historical price data sets Doing by analyzing the patterns in the set; (iii) calculating an estimate for the requested fare price using the estimated fare result of (ii);
(Iv) sending the request to a delivery system for the fare price for the trip from the origin to the destination on a specific date;
(V) receiving from the distribution system a fare price of the distribution system for a trip from a departure point to a destination on a specific date;
(Vi) Compare the calculated estimate for the requested fare price obtained from the calculated estimate of (iii) with the fare price of the distribution system received in (v), for the trip from the origin to the destination on a specific date And (vii) outputting estimated fare class availability for a trip from a departure point to a destination on a specific date. Server is provided.

  In the server, for (ii), the server (a) obtains past fare estimates from a computer data store, (b) groups past fare estimates by category, and (c) Deriving statistical data for each group; (d) storing a classifier that includes the derived statistical data for each group; and (e) using the stored classifier, the requested fare price is It is arranged to identify the corresponding group.

  In the server, for (iii), the server calculates statistical data from the stored classifier corresponding to the identified group to calculate a set of estimates for the requested fare price over the specified date range. It is arranged to perform using.

  The server is further arranged to perform a method including any of the method limitations described above.

  Further, a computer program product embodied on a non-transitory storage medium, wherein the computer program product, when executed on a computer, is for travel from a departure point to a destination on a specific date. Arranged to infer which fare classes are available and when the computer program product is executed on a computer, (i) the fare price for the trip from the origin to the destination on a specific date And (ii) inferring, estimating or predicting estimated fare prices from incomplete historical price data sets by analyzing patterns in the data sets And (iii) an estimate for the requested fare price using the estimated fare result of (ii) And (iv) sending a request to the delivery system for the fare price for the trip from the departure point to the destination on a specific date, and (v) from the departure point to the destination on the specific date. Receiving the fare price of the distribution system for the trip of, and comparing the calculated estimate for the requested fare price obtained from the calculation estimate of (vi) (iii) with the fare price of the distribution system received in (v) Inferring which fare class is available for travel from the origin to the destination on the date of, and (vii) inferred fare class usage for the travel from the origin to the destination on a specific date A computer program product arranged to output the possibilities is provided.

  The computer program product is arranged to perform a method according to any of the method limitations described above.

  A computer connected to a server of any of the foregoing concept B server concepts, wherein the computer is arranged to send a request defining a trip from a starting point to a destination on a particular date; In addition, a computer is provided that is arranged to receive an output inferred fare class availability for a trip from a departure point to a destination on a specific date.

C. Method for Estimating Fare Price A method for estimating a fare price comprising the steps of: (i) obtaining a past fare estimate from a computer data store; and (ii) grouping past fare estimates by category (Iii) deriving statistical data for each group; (iv) storing a classifier containing the derived statistical data for each group on the computer; (v) at a specific date; Receiving a request for a fare price for a trip from a departure place to a destination; (vi) identifying a group to which the requested fare price corresponds using a stored classifier; and (vii) a requested fare price. Statistical data obtained from stored classifiers corresponding to the identified group And using (viii) providing a requested fare price estimate.

D. Method for Estimating Fare Price A method for estimating a fare price, the method comprising: (i) receiving a request for a fare price for a trip from a departure point to a destination on a specific date; and (ii) estimated Inferring, estimating or predicting a fare from an incomplete historical price data set by analyzing patterns in the data set; (iii) using the result of step (ii) to request Calculating an estimate for the fare price determined; and (iv) providing a requested fare price estimate.

E. A method for providing a set of round-trip fare price estimates, comprising: (i) obtaining a past fare estimate from a computer data store; and (ii) a past (Iii) deriving statistical data for each group; and (iv) storing on the computer a classifier including the derived statistical data for each group. And (v) receiving a request for a round trip fare price for a specified departure date and a specified return date for a trip from the origin to the destination, and (vi) requested using a stored classifier Identifying the group to which the round-trip fare price corresponds, and (vii) for the specified departure and return date Calculating a set of estimates for the requested round-trip fare price using statistical data obtained from a stored classifier corresponding to the identified group; and (viii) providing a set of fare price estimates. ,including.

F. A method for providing a set of best round-trip fare price estimates A method for providing a set of best round-trip fare price estimates for a specified departure date range and a specified return date range, comprising: (i) computer data Obtaining past fare estimates from the store; (ii) grouping past fare estimates by category; (iii) deriving statistical data for each group; (iv) each group on the computer And (v) receiving a request for a round trip fare price for a specified departure date range and a specified return date range for a trip from a departure location to a destination. And (vi) using the stored classifier, the group to which the requested round-trip fare price corresponds And (vii) calculating a set of estimates for the requested round-trip fare price for the departure and return date pair using statistical data obtained from the stored classifier corresponding to the identified group; , (Viii) selecting a best fare price estimate for each pair of departure and return dates; and (ix) providing a set of best round-trip fare price estimates for each pair of departure and return dates. .

G. A method for providing a set of round-trip fare price estimates, the method comprising: (i) a specified departure date and a specified return for a trip from a departure point to a destination. Receiving a request for a round-trip fare price for a day; and (ii) analyzing a pattern in that dataset to infer, estimate, or predict an estimated fare from an incomplete historical price dataset. And (iii) using the result of step (ii) to calculate a set of estimates for the requested round-trip fare price for the specified departure and return date; and (iv) providing a set of fare price estimates Including the steps of:

H. A method of providing the best set of round-trip fare price estimates A method of providing the best set of round-trip fare price estimates for a specified departure date range and a specified return date range, the method comprising: (i) Receiving a request for a round-trip fare price for a specified departure date range and a specified return date range for the trip to the destination; and (ii) estimating and estimating the estimated fare from an incomplete historical price data set Or (iii) performing a prediction by analyzing patterns in the data set; and (iii) using the results of step (ii), a set of estimates for the requested round-trip fare price for the departure and return date pair And (iv) provide a set of best round-trip fare price estimates for each pair of departure and return dates Including the steps of:

Claims (45)

  A method for providing a price estimate, comprising: (i) receiving by a computer server a request for parameters defining the goods or services together with prices for the goods or services, eg airfare; and (ii) the aforementioned steps (i Configuring one or more processors to determine an estimated price from an incomplete past price data set at any point in time by analyzing a pattern in the data set; iii) configuring one or more processors to calculate an estimate for the requested price for the goods or services that satisfy the parameters; and (iv) the price estimate for an end user computing device, eg, a personal The method comprising the computer, and sending to the smartphone or tablet, the.   The parameters that define the goods or services include: action type, eg air fare, hotel reservation, rail fare; date range; destination; origin; desired weather condition; star rating; keyword; any other user-defined The method of claim 1, wherein one or more of the preferences are included.   3. A method according to claim 1 or 2, wherein the determination of the estimated price is performed by inferring, deriving or predicting the estimated price.   Step (ii) includes (a) obtaining past price estimates from a computer data store; (b) grouping the past price estimates by category; and (c) statistical data for each group. Deriving; (d) storing on the computer a classifier including the derived statistical data for each group; and (e) using the stored classifier, the requested price corresponds. The method according to claim 1, wherein identifying a group is included.   In step (iii), calculating a set of estimates for the required price over a specified date range is performed using statistical data obtained from the stored classifier corresponding to the identified group. 5. The method of claim 4, wherein   The method according to any of claims 1 to 5, wherein step (ii) comprises analyzing a pattern in the data set using rules.   The method according to any of claims 1 to 6, wherein step (ii) includes a naive Bayes classifier machine learning approach that generates a probability model of a price and uses that model to predict an unobserved price.   The method of claim 7, wherein the classifier is trained using observed prices and their corresponding feature sets.   9. The method of claim 8, wherein the features relate to the request and include one or more of a weekly departure date, number of days stayed, stay on Saturday, airline, time to travel, route, month.   9. The method of claim 8, wherein the classifier then predicts the price of the unobserved price by being given a set of features and giving a price that is most likely to have those features. .   The method of claim 8, wherein deriving a feature may be performed by training a plurality of models with different features and comparing the prediction accuracy of the different models.   Step (ii) includes building a statistical model from past prices, identifying missing quotation candidates, and pricing quotation candidates based on the statistical models Item 8. The method according to any one of Items 1 to 7.   Estimating the price for each candidate estimate, extracting a category feature value from the estimate, extracting a trained classifier from the database for the extracted category, and all from the estimate candidate Extracting the feature value of the class, classifying the candidate estimate, calculating a Bayesian posterior probability for each price range stored in the classifier, and a price range class with the highest Bayesian posterior probability The method of claim 12, wherein the method comprises: selecting and assigning a price class to the candidate quote.   14. A method according to claim 12 or 13, wherein the inputs to the statistical model include a list of paths, a classifier category classification scheme, a past estimate, and a set of supported features with weights.   The method of claim 14, wherein past estimates are filtered by age.   14. A method according to claim 12 or 13, wherein the input to the statistical model includes a reverse path equivalent.   17. A method as claimed in any preceding claim, including the step of including a cached fare price in the set of price estimates.   18. A method as claimed in any preceding claim, wherein the set of price quotes is constructed without querying the distribution system after step (i).   19. A method according to any preceding claim, wherein the price is for a trip that is a one-way trip.   19. A method according to any preceding claim, wherein the price is for a trip that is a round trip.   21. A method as claimed in any preceding claim, wherein the price includes an air fare price.   The method according to any one of claims 1 to 21, wherein the price includes a railway fare price.   The method according to any one of claims 1 to 22, wherein the price includes a rental car price.   24. A method according to any preceding claim, wherein the price includes a hotel price.   25. A method as claimed in any preceding claim, wherein the request includes a flexible search request.   26. A method according to any preceding claim, wherein the final result of the method is at most one estimate per input date or input date pair.   (A) At any point in time for step (i), determining the confidence range of the estimated price from an incomplete past price data set is performed by analyzing the pattern in the data set or Configuring a plurality of processors; and (B) configuring the one or more processors to calculate a confidence range of the estimated price for the required price for the goods or services that satisfy the parameters. 27. A method according to any of claims 1-26.   28. The method of claim 27, further comprising: (C) providing the fare price estimate along with the confidence range to an end user computing device, such as a personal computer, smartphone, or tablet.   28. The method of claim 27, further comprising using the confidence range to determine whether to display a price to the user or to provide a likely price range to the user.   30. The method of claim 29, wherein the likely price range is displayed as an error bar.   25. A method according to any preceding claim, wherein the date range includes only one departure date.   25. A method according to any of claims 1 to 24, wherein the date range includes only one return date.   The method includes providing a set of best round-trip fare price estimates for a specified departure date range and a specified return date range, and step (i) includes specifying for a trip from the origin to the destination. Receiving a request for a round trip fare price for a specified departure date range and a specified return date range, wherein step (iv) includes the best round trip fare price for the specified departure date range and the specified return date range. 31. A method according to any preceding claim, including providing a set of estimates.   34. The method of claim 33, wherein the set of best round-trip fare price estimates is provided in a bar chart format.   One of the minimum Bayesian posterior probabilities necessary for the estimation process to accept the classification result, the maximum number of path operators involved in candidate generation, or a random variation added to the Bayesian posterior probabilities to avoid ties 35. A method according to any of claims 1 to 34, or parameterized by a plurality.   36. The method according to any of claims 1 to 35, wherein the method is performed on a server that provides a fare availability and price quote service.   A server configured to provide a price quote, wherein: (i) receiving a request for parameters defining the goods or services together with prices for the goods or services, such as airfare; Receiving, including one or more of any other user-defined preferences, such as airfare, hotel reservation, rail fare; date range; destination; origin; desired weather conditions; star rating; (Ii) determining an estimated price from an incomplete past price data set at any point in time for (i) above by analyzing a pattern in the data set; (iii) Estimate for the requested price for the goods or services that meet the parameters And calculating, arranged to perform, and to provide the price quote (iv) server.   For (ii), (a) obtaining past prices from a computer data store; (b) grouping the past prices by category; and (c) deriving statistical data for each group. And (d) for each group, storing a classifier including the derived statistical data, and (e) identifying a group using the stored classifier to which the required price corresponds. 38. The server of claim 37, arranged to perform.   For (iii), calculating a set of estimates for the required price over a specified date range is performed using statistical data obtained from the stored classifier corresponding to the specified group. 40. The server of claim 38, located in   40. A server according to any of claims 37 to 39, further arranged to perform the method according to any of claims 6 to 36.   A computer program product embodied on a non-transitory storage medium, wherein the computer program product is arranged to provide a price quote when executed on a computer, the computer program product When executed on a computer, (i) receives a request for parameters defining the goods or services along with prices for the goods or services, such as airfare, where the parameters include an action type, such as airfare, Hotel reservation, rail fare; date range; destination; origin; desired weather conditions; star rating; keyword; including one or more of any other user-defined preferences, receiving (ii) From an incomplete historical price data set at any point in time for (i) Determining an accumulated price by analyzing patterns in the data set; (iii) calculating an estimate for the required price for the goods or services that satisfy the parameters; and (iv) the price. A computer program product arranged to provide and provide a quote.   42. A computer program product according to claim 41 arranged to perform the method according to any of claims 2-36.   41. A computer terminal connected to a server according to any of claims 37 to 40, wherein the computer terminal sends a request for a price quote for a specified date range for a trip from a departure place to a destination. A computer terminal, wherein the computer terminal is further arranged to receive a price quote.   An article or service, such as an air ticket, for which a price estimate for the article or service is provided using a method according to any of claims 1-36 or a server according to any of claims 37-41. Goods or services.   A web-based price comparison web site where end users can interact with the web-based price comparison web site by providing a price request for goods or services; 37. A web-based price comparison web site that provides the request to a server that estimates the price using the method of any of claims 1-36.