TWI852329B - Object recommendation device based on knowledge level - Google Patents

Abstract

The object recommendation device based on the degree of recognition of the present invention mainly includes an object database, a buyer database, a search module, and a transmission module. The search module receives the event notification that the sales status of the object has been changed within a predetermined time, and only when the behavior attribute of the buyer meets a recommendable threshold value, it finds multiple buyer data that match the conditions of the object whose sales status has been changed according to the content defined by the viewed object in the behavior attribute of the buyer over time, and generates a recommended buyer list containing a specific amount of multiple buyer data. The transmission module enables the display device of the client device to present the recommended buyer list.

Description

Object recommendation device based on knowledge level

The present invention relates to an object recommendation device, in particular to an object recommendation device based on a withdrawal.

In addition to community bulletin boards or physical service stores, real estate websites can provide detailed information on real estate sales or rentals nationwide and can be combined with personalized services. Therefore, website information technology has become a goal that real estate developers have been working hard to develop and develop in recent years to enhance their competitiveness.

In order to improve the convenience of users when using home buying websites, most home buying websites provide the function of user favorites. When browsing and searching for housing objects, users can click the "Favorites" button or similar function buttons on the object display page to add their favorite objects to their favorites. After that, users can use their favorites to check, track, compare or purchase the objects they like or are interested in on a daily basis.

When entering the transaction negotiation process, in order to protect the rights and interests of both parties, as long as a buyer makes an offer, the sales status of the object in the transaction negotiation will be changed to mediation, and other buyers will be prohibited from making bids. After the current buyer and seller reach or fail to reach a transaction, the sales status will be changed to non-sales status, and other buyers will be allowed to bid again to avoid bidding by multiple buyers. This is the so-called single mediation system. However, if the sales status is changed to the non-sales status too slowly, it may cause damage to the interests of the seller. Therefore, some businesses have changed the sales status to an online one to improve the efficiency of real-time updates. However, this still does not provide enough protection for the interests of the seller and may easily lead to the loss of the seller's customers.

In addition, if you make a recommendation without knowing enough about the buyer, it is easy to make an inaccurate recommendation.

Therefore, the applicant, in view of the lack of knowledge and technology, invented the "object recommendation device based on cognitive level" in this case to improve the above-mentioned shortcomings.

In order to overcome the problems existing in the knowledge and technology, the present invention provides an object recommendation device based on the degree of recognition of the present invention, which mainly includes an object database, a buyer database, a search module, and a transmission module. The search module receives the event notification that the sales status of the object has been changed within a predetermined time, and only when the behavior attribute of the buyer meets a recommendable threshold value, it finds multiple buyer data that match the conditions of the object whose sales status has been changed according to the content defined by the buyer's behavior attribute that changes with time, and generates a recommended buyer list containing a specific number of multiple buyer data. The transmission module enables the display device of the client device to present the recommended buyer list. In this way, the user only needs to simply select a specific behavior to allow the system to search for the corresponding object for selection.

The object recommendation device based on the degree of recognition provided by the present invention proposes a method that can quickly match the rejected object based on the objects browsed by the buyer after the rejection, so as to quickly find other objects worth seeing, avoiding the user to spend a lot of time and energy to find these recommended objects, while protecting the rights and interests of the seller.

10: Object recommendation device based on level of familiarity

101: Object database

100: Buyer database

102: Recording space

103: Recording module

105:Transmission module

106: Presentation module

107:Search module

5: Client equipment

20: Operation screen

202: First mark button

204: Collection function button

206: Favorite button for a single object

208: Subscription function button

22: Display block

The first figure shows a system schematic diagram of a specific embodiment of the object recommendation device based on the degree of recognition proposed in the present invention.

The second figure shows another schematic diagram of a specific embodiment of the object recommendation device based on the degree of recognition proposed in the present invention.

The third figure shows a schematic diagram of the operation screen of the present invention.

The present invention will be further described in detail through the following preferred embodiments and their accompanying drawings.

Referring to the first figure, the object recommendation device based on the degree of recognition of the present invention mainly includes an object database 101, a buyer database 100, a search module 107, and a transmission module 105. The object database 101 stores a plurality of object data, each of which has an object index code and a plurality of object attributes to which the object index code belongs, and the object attributes at least include the sales status in the transaction negotiation.

The client device 5 transmits data to and from the object recommendation device 10 based on the degree of recognition of the present invention through the Internet. Specifically, the presentation module 106 transmits the screen to be presented to the client device for presentation, and the information input by the user in the various screens presented by the presentation module 106 is first received and further transmitted to the relevant module, but this does not mean that the object recommendation device 10 based on the degree of recognition of the present invention must include the presentation module 106. The aforementioned client device 5 can be a desktop computer, a laptop, a tablet computer or a smart phone, etc.

Each object data in the object database 101 has an object index code and a plurality of object attributes to which the object index code belongs. Among them, the object attributes include the abbreviation, price, community name, address, floor, building registration area, land registration area, unit price per unit area, type, layout, house age, parking space, orientation, elevator, management fee, layout map, and living function of the real estate object. Each of the plurality of object attributes has a priority, and the higher the priority, the more highly overlapped the priority will be, which will affect the objects presented in the recommended buyer list.

The buyer database 100 stores multiple buyer data, each of which has a behavior index code and multiple behavior attributes to which the behavior index code belongs, and each behavior attribute is mainly defined by the objects that have been viewed over time. Since the system can first label all the objects and buyer requirements, for example, if the object belongs to three generations living under one roof and the buyer's requirement is three generations living under one roof, matchmaking operations can be performed through labeling.

In brief, the search module 107 receives an event notification that the sales status of the object has been changed within a predetermined time, and only when the behavior attribute of the buyer meets a recommendable threshold, it finds multiple buyer data that match the conditions of the object whose sales status has been changed according to the content defined by the buyer's behavior attribute that changes with time, and generates a recommended buyer list containing a specific number of multiple buyer data, and the transmission module 105 enables the display device of the user 5 to present the recommended buyer list. The financial conditions include self-prepared funds, monthly affordable mortgage, loan term, and number of existing mortgages. Among them, only when the behavior attribute of the buyer meets a recommendable threshold value, the content defined by the viewed object in the behavior attribute of the buyer changes with time, so as to avoid making recommendations without sufficient knowledge of the buyer, which is easy to be inaccurate, and at the same time, it can prevent the old buyer behavior from excessively interfering with the current buyer behavior.

The behavior attribute of the buyer meeting the recommendable threshold is related to the degree of familiarity with the buyer, and the degree of familiarity can be defined as the objects viewed within the predetermined time reaching the threshold. In other words, the degree of familiarity can be defined as the degree to which the objects viewed within the predetermined time reach the threshold and the object attributes of the objects are identifiable from each other. Theoretically, if the properties viewed by the buyer lack a certain degree of concentration, for example, if he has viewed 10 properties but they are located in 8 administrative regions, it means that the buyer has not yet determined in which administrative region the property he wants to buy is located. If we rashly start recommending properties to the buyer at this time, the recommended properties may easily be located in an administrative region that the buyer is not interested in, thus becoming an invalid recommendation.

In other words, in addition to calculating the total price range that the buyer can purchase based on the financial conditions, the search module 107 can further label the property and buyer requirements based on the property attributes, such as the administrative area and the house type, in the housing purchase requirements, and reasonably reduce the asking price of the property based on the appropriate bargaining rate, taking into account the differences in house viewing behavior and changes over time, and then reasonably find multiple buyer information that meets the financial conditions and provide the recommended buyer list.

In order to improve the accuracy of the recommended buyer list, the recommended buyer list can also take into account the user's browsing behavior to avoid too many items, resulting in the user still being unable to find his ideal item.

Referring to the second figure, it shows a system schematic diagram of a specific embodiment of the object recommendation device based on the degree of recognition of the present invention. The object recommendation device based on the degree of recognition 10 and the object database 101 storing multiple object data can transmit messages to each other through unspecified communication means, and the transmission device 10 that can provide a recommended buyer list includes at least a recording space 102 and a recording module 103 for receiving user input from the client device 5 for recording browsing behavior. The recording space 102 is configured to store multiple browsing records, each browsing record is the object index code corresponding to the user code.

In short, when a user uses the client device 5 to browse the object provided by the system to view the website, the recording module 103 will detect and receive the browsing event from the client device. Since the browsing event includes the user code and the object index code, the object data corresponding to the user code and the object index code in the browsing event can be collected in the recording space 102. As long as the collected browsing events reach a certain level and a meaningful recommended buyer list can be generated, as long as the search module 107 receives an event notification that the sales status of the object has been changed within a predetermined time, in addition to searching based on the content defined by the viewed objects in the buyer's behavioral attributes that change over time only when the buyer's behavioral attributes meet a recommendable threshold, it can also summarize the object index codes in the multiple browsing records based on the multiple browsing records to which the user code belongs, and the object attributes with high overlap, and use the transmission module to transmit the recommended buyer list to the client device 5 to which the user code belongs. Among them, the recommended buyer list is at least one object index code related to the highly overlapping object attributes.

Please refer to the third figure. When the user uses the client device 5 to browse the object provided by the system to view the website, the search results entered by the user will be listed on the client device 5. As long as the user clicks on an object, he will trigger a browsing event, so that the recording module 103 will store the object data corresponding to the user code and the object index code in the browsing event into the recording space 102.

Since each object data in the object database 101 has an object index code and multiple object properties to which the object index code belongs, the search module 107 can use the object index code of each browsing record stored in the record space 102 to find the multiple object properties to which the object index code belongs, and then summarize the object properties with high overlap for the object index codes in multiple browsing records. For example, if the user frequently and repeatedly browses building objects in the same area, the search module 107 determines that the user has a high interest in building objects in a specific area, and lists the building objects in the specific area that have not been browsed as having highly overlapping object attributes, and includes them in the recommended buyer list, such as the specially recommended display block 22 in the third figure. However, since the number of building objects that can be included in the recommended buyer list is limited, such as 3 items, the search module 107 will need to perform further screening as long as the number of building objects in a specific area is too large. That is, expand more attributes of the object to find objects with high overlap. For example, in addition to building objects in the same area, users often browse objects with nearby parks, hospitals, subway stations, etc.

However, in order to make the objects included in the recommended buyer list closer to the user's expectations, each of the multiple object attributes has a priority, and the higher the priority, the more overlapped the priority. In other words, for example, if the user frequently clicks on objects that have parks and hospitals but no MRT stations in the recommended buyer list, even if the object attribute of the object index code in multiple browsing records often contains the MRT station object attribute, the priority of the MRT station object attribute will be reduced, and the possibility of being included in the recommended buyer list in the future will also be reduced. Below, further explanation is given on how to list the search results entered by the user on the client device 5.

Referring to the second figure, another system schematic diagram of a specific embodiment of the object recommendation device based on the degree of recognition of the present invention is shown. The object recommendation device based on the degree of recognition 10 and the object database 101 storing multiple object data can communicate with each other through unspecified communication means, and the transmission device 105 that can provide a list of recommended buyers at least includes a recording space 102, a search module 107 for receiving user input from the client device 5 for searching and comparing object data in the database 101, a presentation module 106 for presenting various screens, and a recording module 103 for recording browsing behavior.

The search conditions are input through the search screen 20 presented by the presentation module 106, and the input method can be to click on the menu or to type on the keyboard. The search conditions are first received from the client device 5 through the Internet, and the search conditions are transmitted to the search module 107. The search module 107 compares the search conditions with the plurality of object data in the database 101, thereby generating a search result, which includes a plurality of expected object data that match the search conditions and a plurality of object index codes corresponding to the plurality of expected object data, and each expected object data corresponds to a unique object index code. In a preferred specific embodiment, each object has an object number that is not repeated with other objects as its unique object index code.

For example, when a user enters the search condition "New Taipei City, Yonghe District, and housing between NT$8 million and NT$12 million" through the search screen 20 and presses the search button to issue a search event, the search module 107 receives the search event including the search condition, compares the search condition with the plurality of object data in the database 101, and generates a search result. As shown in FIG. 3 , the presentation module 106 presents the plurality of desired object data matching the search condition through the operation screen in a system-default sorting order, a first marking button 202 corresponding to each desired object data, and a collection function button 204 for collecting the marked object. The user can mark the house object data of interest through the operation screen (usually, the user will mark multiple house object data), and collect all the marked object data at the same time by clicking the collection function button 204. When the collection function button 204 is activated, a collection event will be issued, and the collection event includes basic information of the marked object data, and the basic information at least includes the object index code of the marked object data, and may also include, for example, the collection time, object name, object address, etc. The user can also click the collection button 206 for a single object to trigger the collection event of the single object.

The recording module 103 detects at any time whether a collection event occurs. When the recording module 103 detects and receives a collection event, the basic information of several pieces of object data in the collection event is simultaneously collected in a recording space 102 corresponding to the user, so that the user can review the collected objects in the recording space 102 when logging into the website. The user's respective recording space 102 can be set in the database 101 or independently of the database 101, and can also be in the form of an object index code list or list. For the sake of convenience, the term "collection folder" is used below to represent each user's dedicated recording space 102 or collection list. For users who need to quickly collect a large amount of housing object information, a system that can collect multiple pieces of object data at the same time is quite attractive to them. Even as shown in the third picture, users can check the "Select All" button to save all the object data found with specific search criteria to the favorites folder at once.

After the user enters the search condition "New Taipei City, Yonghe District, and residential properties between NT$8 million and NT$12 million" through the search screen 20, if the user cannot find the object of interest from the currently displayed object data, the user can press the subscription function button 208 in the operation screen that will be displayed after the search key is pressed, and let the system help track it. When the subscription function button 208 is activated, a subscription event will be issued, and the subscription event at least includes the search condition "New Taipei City, Yonghe District, and residential properties between NT$8 million and NT$12 million". The recording module 103 detects at any time whether a subscription event occurs. When the recording module 103 detects and receives a subscription event, it can respond to the subscription event and execute the subscription procedure.

The present invention is disclosed with reference to the above-mentioned preferred embodiments and examples. The reader should understand that these examples are used for description rather than limitation. Anyone who is familiar with this art can make various combinations and modifications without departing from the spirit and scope of the present invention, which should still fall within the scope of the patent of the present invention.

10: Object recommendation device based on level of familiarity

100: Buyer database

101: Object database

105:Transmission module

106: Presentation module

107:Search module

5: Client equipment

Claims (8)

An object recommendation device based on recognition level includes: an object database storing a plurality of object data, each object data having an object index code and a plurality of object attributes to which the object index code belongs, and the object attribute at least includes a sales status in transaction negotiation; a buyer database storing a plurality of buyer data, each buyer data having a behavior index code and a plurality of behavior attributes to which the behavior index code belongs, and each behavior attribute is mainly defined by the viewed object changing with time; a search module coupled to the object database and the buyer database, It is configured to receive an event notification that the sales status of the object is changed within a predetermined time, and only when the behavior attribute of the buyer meets a recommendable threshold value, it finds multiple buyer data that match the conditions of the object whose sales status is changed according to the content defined by the viewed object in the behavior attribute of the buyer over time, and generates a recommended buyer list containing a specific number of multiple buyer data; and a transmission module, coupled to the search module, is configured to enable a display device of a client device to present the recommended buyer list. As described in item 1 of the patent application scope, the object recommendation device based on the degree of recognition, wherein the object attributes include a short name, price, community name, address, floor, building registration area, land registration area, unit price per unit area, type, layout, age, parking space, orientation, elevator, management fee, layout map, and living function of a real estate object. As described in item 2 of the patent application scope, an object recommendation device based on recognition level, wherein each of the multiple object attributes has a priority, and the higher the priority, the more highly overlapping the priority. The object recommendation device based on the degree of recognition as described in item 1 of the patent application scope further includes: the search module is configured to receive a search condition, compare the search condition with the plurality of object data in the object database, and thereby generate a plurality of expected object data that match the search condition and a plurality of object index codes corresponding to the plurality of expected object data, wherein the search condition is received from a client device through the transmission module. As described in item 4 of the patent application scope, the object recommendation device based on the degree of recognition, the client device further includes: a presentation module, configured to present through an operation screen a plurality of desired object data belonging to a user code and matching the search condition, a first marking button corresponding to each desired object data, and a collection function button for collecting the object index code corresponding to the marked desired object data, wherein the collection function button will issue a collection event when it is activated. As described in item 1 of the patent application scope, the device for recommending objects based on the degree of recognition, wherein the behavior attribute of the buyer satisfies the recommendable threshold value and is related to the degree of recognition of the buyer. As described in item 6 of the patent application scope, the object recommendation device based on the degree of recognition, wherein the degree of recognition can be defined as the object viewed within the predetermined time reaches a threshold value. As described in item 7 of the patent application scope, the object recommendation device based on the degree of recognition, wherein the degree of recognition can be defined as the object viewed within the predetermined time reaches a threshold value and the object attributes of the object have a recognizable degree of concentration between each other.

Images