TWI647578B - Search engine based document indexing method, data query method and server - Google Patents

Abstract

The embodiment of the present application discloses a document indexing method based on a search engine, a data query method, and a server. The document indexing method includes: acquiring a document to be indexed, and performing a word segmentation operation on the document to obtain a unitary word segmentation; Whether the unary participle is a filter word, if the unary participle is a filter word, the unary participle and the at least one unary participle adjacent to the unary participle sequence are into a multivariate participle, and the multivariate participle is indexed; If the unary participle is not a filter word, the index of the unary participle is directly indexed. In the embodiment of the present application, when indexing or querying, the unary word segmentation as a high frequency word and its adjacent at least one unary word segmentation are multi-particials to ensure that the query is not wasted due to indexing the high frequency word. Search engine resources, and will not cause inaccurate query results by skipping the index of high frequency words.

Description

Search engine based document indexing method, data query method and server

The application relates to the field of search engine technology, and particularly relates to a document indexing method based on a search engine, a data query method and a server.

Search Engine refers to the use of specific computer programs to organize and process information according to certain strategies and information, and display the processed information to users to provide users with search services. system.

The search engine works as follows: First, web crawling, each independent search engine has its own web crawler, commonly known as web spider (Spider), Spider follows the hyperlink in the web page, continuously grabbing Take the webpage, the webpage that is captured is called the webpage snapshot. Since the application of the hyperlink in the internet is very common, in theory, from a certain range of webpages, most of the webpages can be collected; secondly, proceed Web page processing, after the search engine catches the webpage, extracts the keywords and builds the index file; in order to provide the retrieval service; finally, provides the retrieval service, the user inputs the keyword to search, and the search engine finds the webpage matching the keyword from the index database. In addition to the page title and URL, a summary of the page and other information will be provided for the user's convenience.

For Chinese search engines, Chinese indexing operations are required for indexing and querying. The commonly used Chinese word segmentation method is the unary word segmentation method, which means that each Chinese character in the sentence is treated as a unit. The indexed sentence is “China Stock Market”, and the result after the one-yuan participle is four words, namely “中中”, “国”, “股”, “市”. Taking the word "city" as an example, in a single search engine server indexing 6 million documents, the probability of "city" appearing is as high as 93%. Therefore, when querying "Chinese stock market" based on the result of dividing the unitary word, The query of "city" will greatly consume the resources of the search engine server. Therefore, in the search engine, the list of high-frequency words is pre-stored, and the high-frequency words are not filtered, so searching for "Chinese stock market" is simplified. To search for "China stocks", to skip the query for the high-frequency word "city".

In the research and practice of the prior art, the inventors found that the prior art has the following problems: when using the unary word segmentation for indexing and querying, although the query of the high frequency word is skipped through the preset high frequency word list , but it will lead to inaccurate query results. Still taking the query "China stock market" as an example, although the query of "city" is skipped, the returned query results will include a large number of "Chinese stocks", "Chinese stocks" and other query results containing "Chinese stocks". As a result, the result of the query does not match the content that needs to be queried.

The purpose of the embodiment of the present application is to provide a document indexing method based on a search engine, a data query method, and a server, so as to solve the problem that the existing indexing and querying by the high frequency word filtering method is performed, resulting in inaccurate query results.

In order to solve the above technical problem, an embodiment of the present application provides a The search engine-based document indexing method is implemented as follows: a search engine-based document indexing method includes: acquiring a document to be indexed, and performing a word segmentation operation on the document to obtain a unary word segmentation; determining whether each unary segmentation word is filtered a word, if the unary participle is a filter word, the unary participle and the at least one unary participle adjacent to the unary participle sequence are multivariate word segmentation, and the multivariate participle is indexed; if the unary participle is not a filter word , directly indexing the unary participle. To solve the above technical problem, an embodiment of the present application provides a data query method based on a search engine, which is implemented by a search engine-based data query method, and the data query method is established by applying the document index method. The index includes: obtaining the data to be queried, and performing a word segmentation operation on the data to obtain a unary word segmentation; determining whether each unary segmentation word is a filter word, and if the unary segmentation word is a filter word, the unitary word segmentation and the Locating at least one unary word segment adjacent to the unary word segmentation into a multivariate word segment, searching for the index according to the multivariate word segment, if the unary segmentation word is not a filter word, searching for the index according to the unary segmentation word; The resulting query results are merged.

To solve the above technical problem, an embodiment of the present application further provides a search engine-based document index server, which is implemented as follows: a search engine-based document index server, including: An obtaining unit, configured to obtain a document to be indexed; a word segmentation unit, configured to perform a word segmentation operation on the document obtained by the obtaining unit to obtain a unary word segmentation; a determining unit, configured to determine whether each unary segmentation word is a filtering word; When the judging unit judges that the unary participle is a filter word, the unary participle and the at least one unary participle adjacent to the unary participle sequence are into a multi-word segmentation, and the multi-particial word is indexed when When the judging unit judges that the unary participle is not a filter word, it directly indexes the unary participle.

In order to solve the above technical problem, an embodiment of the present application further provides a data query server based on a search engine, which is implemented by: a search engine based data query server, wherein the data query server applies the document The index established by the index server includes: an obtaining unit for acquiring data to be queried; a word segment unit for performing a word segmentation operation on the data obtained by the obtaining unit to obtain a unary word segmentation; and a determining unit for determining each one dollar Whether the word segmentation is a filter word; the searching unit is configured to, when the determining unit determines that the unary word segmentation is a filter word, the plurality of word segmentation words and at least one unary word segment adjacent to the unitary word segmentation sequence into a multi-word segmentation, according to The multivariate word segmentation searches the index, and when the determining unit determines that the unary segmentation word is not a filter word, the index is searched according to the unary word segmentation; the merging unit is configured to search the index unit after searching the index The obtained query results are merged.

It can be seen that, in the embodiment of the present application, when indexing or querying, the unary word segmentation of the high frequency word and the adjacent at least one unary word segmentation are multivariate word segmentation, so as to ensure that the query is not caused by indexing the high frequency word. It wastes the resources of the search engine, and does not cause the query result to be inaccurate because of skipping the index of the high frequency word; taking the example of "China stock market" as an example, the embodiment of the present application will use the word "zhong" and " The country, the "stock", and the "stock market" are queried. The hit rate of the segmentation "stock market" is much lower than the hit rate of the high-frequency word "city", and it will not be queried by skipping the query for "city". In addition to the "stock market" word segmentation, it can reduce the search engine resource consumption while returning the correct query results, thereby improving the performance of the search engine.

An embodiment of the present application provides a document indexing method based on a search engine, a data query method, and a server.

In order to make those skilled in the art better understand the technical solutions in the embodiments of the present application, and the above objects, features and advantages of the embodiments of the present application can be more clearly understood, the present application will be described below with reference to the accompanying drawings. The technical solutions in the embodiments are described in further detail.

The search engine is divided into four parts: a search module, an index module, a query module, and a user interface module. Among them, the function of the search module is to roam through the Internet through the Spider to discover and collect webpage information; the function of the index module is to extract index items from the webpages searched by the search module to represent the documents and generate documents. Library index table; check The function of the query module is to retrieve documents in the index library according to the user's query, and to sort the results to be output, and to feed back the information according to the user's query requirements; the function of the user interface module is to receive the query request of the user, and The user returns the result of the query. The embodiment of the present application mainly describes an implementation process of an index function and a query function in a search engine.

In order to enable a person skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present application. It is apparent that the described embodiments are only a part of the embodiments of the present application, and not all of them. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope should fall within the scope of the protection of the present application.

Referring to FIG. 1, a flowchart of a first embodiment of a search engine-based document indexing method according to the present application is as follows:

Step 101: Acquire a document to be indexed.

The document to be indexed in the embodiment of the present application is usually webpage information collected from the Internet through a search function of a search engine, and the webpage information is stored in a storage space of a search engine server (for example, a magnetic disk). When the search engine server needs to index, it obtains the information of the webpage that has not been indexed from the storage space.

Step 102: Perform a word segmentation operation on the indexed document to obtain a unary word segmentation.

It should be noted that if there is only one word in the document to be indexed, and the word is a high frequency word, the document is no longer indexed.

Step 103: Determine whether each unary participle is a filter word, and if yes, execute step 104: Otherwise, perform step 105.

Among them, the filter word is the high-frequency word that the search engine will ignore directly when querying. These high-frequency words have a high frequency in the document, so the query will consume a lot of system resources, so usually through statistical means. The filter word list is determined before the index, and the query for these high frequency words is skipped for subsequent query. For example, according to the statistical results, the “city” in the “Chinese stock market” is a high-frequency word, so the “city” can be written into the filter word list.

Usually, the unary word segmentation result obtained after the document segmentation includes several unary segmentation words, so the order is judged whether each unary segmentation word is a filter word.

Step 104: The unary participle and the at least one unary participle adjacent to the unary participle are into multivariate word segmentation, index the multivariate participle, and end the current process.

Wherein, preferably, the unary participle and the unary participle adjacent to the unary participle are into a binary participle, and for the current unary participle judged as the high frequency word, if the unary participle is the first unary participle in the document , the binary participle is merged with the latter one-part participle into a binary participle; if the unary participle is the last unary participle in the document, the unary participle is combined with the previous unary participle into a binary participle; if the unary participle is not The first unary participle and the last unary participle in the document respectively form the binary participle with the previous unary participle and the latter unary participle.

For the indexing of binary word segments composed in the above manner, it can be seen from the above description that two adjacent unary word segments are formed into one binary word segment. For example, "I" and "I" are two unary word segments, and then they are combined. The composition of "my" is a binary participle.

Similarly, multivariate word segmentation refers to at least two adjacent unary word segments into a multivariate word segmentation. In addition to the above binary word segmentation, there are also ternary participles, quaternary participles, etc., for example, by "middle", "national" and "person" Form a ternary participle "Chinese."

Step 105: Directly index the unary participle and end the current process.

For unary words that are not high frequency words, they are directly indexed according to the manner in the prior art.

2 is a flowchart of a second embodiment of a search engine-based document indexing method according to the present application. The embodiment describes a document indexing process in detail by taking a multivariate word segmentation as a binary word segmentation as an example:

Step 201: Set a filter word list in advance.

The setting of the filter word list can be calculated by statistical calculation of a large number of documents in the Internet. For example, suppose that 6 million documents are separately scored, then count the number of occurrences of these unary words in each document, and finally sort the number of occurrences of these unary parts, and take the highest ranked (for example, the top 10) unary words. As a high frequency word, a list of high frequency words is constructed.

Step 202: After loading the filter word list, obtain the document to be indexed.

Step 203: performing a word segmentation operation on the indexed document to obtain a unitary score word.

Step 204: Traverse the unitary word segmentation.

Traversing the unary participle, that is, obtaining each unary participle in the order of the word segmentation result of the document, and performing a judging step of whether or not the subsequent filtering word is performed for each of the obtained unary participles.

Step 205: Determine whether the current unary participle is a filter word by searching the filter word list. If yes, execute step 206; otherwise, perform step 208.

Step 206: Combine the unary participle and the unary participle adjacent to the unary segmentation into a binary participle.

Step 207: Index the binary partifier, and perform step 209.

Step 208: directly index the unary participle.

Step 209: Determine whether the unigram is completed. If yes, the current process ends; otherwise, return to step 204.

In the second embodiment of the search engine-based document indexing method of the present application, steps 204 to 208 may be implemented by using the following virtual code:

In the index created by the embodiment of the above document indexing method applied in the search engine, there is no index for the high frequency word, but an index for the binary word segmentation of the high frequency word and its adjacent words is included, which is followed. The accuracy of the data query provides a guarantee.

Corresponding to the embodiment of the search engine-based document indexing method of the present application, the present application further provides an embodiment of a search engine-based data query method, and the embodiment of the data query method uses the document indexing method. The index established by the embodiment performs data query.

3 is a flowchart of a first embodiment of a search engine-based data query method according to the present application;

Step 301: Acquire data to be queried.

The information to be queried is usually the data that the Internet user inputs from the front end of the website, and the search engine receives the input query data.

Step 302: Perform a word segmentation operation on the data to be queried to obtain a unary word segmentation.

Step 303: Determine whether each unary participle is a filter word. If yes, go to step 304; otherwise, go to step 305.

Usually, the unary word segmentation result obtained after the data segmentation includes several unary segmentation words, so the order is judged whether each unary segmentation word is a filter word.

Step 304: Form a unitary participle and at least one unary participle adjacent to the unitary participle into a multivariate participle, according to the multivariate participle search Index, go to step 306.

Preferably, the unary participle and the unary participle adjacent to the unary word segmentation are binary word segments, and for the current unary participle judged as a high frequency word, if the unary participle is the first in the to-be-queried data a one-yuan participle, the binary participle is converted into a binary participle with the latter one-part participle; if the unary participle is not the first unary participle in the to-be-queried material, the unary participle is compared with the previous unary participle or the latter unary The word segmentation is a binary participle.

Step 305: Find an index established according to the unary participle.

Step 306: Combine the query results obtained after the index is searched, and end the current process.

Referring to FIG. 4, it is a flowchart of a second embodiment of a search engine-based data query method according to the present application. The embodiment describes a data query process by taking a multivariate word segmentation as a binary word segmentation as an example.

Step 401: After loading the preset filter word list, obtain the data to be queried.

Step 402: Perform a word segmentation operation on the data to be queried to obtain a unary word segmentation.

Step 403: Traverse the unitary word segmentation.

Traversing the unary participle, that is, obtaining each unary participle according to the word segmentation result of the data to be queried, and performing a judging step of whether or not the subsequent unfiltered word is performed on the obtained unary participle.

Step 404: Determine whether the current unary participle is a filter word by searching the filter word list, and if yes, perform step 405; otherwise, perform steps 407.

Step 405: Combine the unary participle and the unary participle adjacent to the unary segmentation into a binary participle.

Step 406: Perform step 208 according to the index established by the binary participle search.

Step 407: Find an index established according to the unary participle.

Step 408: Determine whether the unitary word segmentation is traversed. If yes, execute step 409; otherwise, return to step 403.

Step 409: Combine all the query results obtained after the index is searched, and end the current process.

For the merging of the results obtained by querying the index according to each word segment, the result of the final display may be returned to the user according to the pre-set conditions (for example, returning 20 results), which is consistent with the prior art, and will not be described herein. .

In the second embodiment of the search engine-based data query method of the present application, steps 403 to 407 can be implemented by using the following virtual code:

In the search engine, the above data query method embodiment is applied to perform data query. Since there is no high frequency word in the established index, and the high frequency word is combined with other words to form a binary word segmentation, an index is established, so the data query is not because Indexing high frequency words wastes the resources of the search engine and does not result in inaccurate results by skipping queries for high frequency words.

In addition, it should be noted that when the index and the query embodiment of the present application are applied to a Chinese search engine, the unary word segmentation obtained by the word segmentation operation refers to each word, and the “Chinese stock market” is taken as an example, according to The filter word preset in the statistical result is “City”, and the undivided word segmentation is “Medium”, “Country”, “Share”, “City”; when the index and query embodiment of the above application is applied to the foreign language search engine In the English "Chinese Stock Market" as an example, it is assumed that according to the statistics "Stock" is a filter word, the unary participles classified according to the characteristics of the English participle are "Chinese", "Stock", "Market", subsequent indexes and queries. The process is consistent with the Chinese characters, that is, when indexing, after filtering "Stock", the indexes obtained are "Chinese", "Chinese Stock", "Stock Market" and "Market"; when querying, you can "Chinese" , "Chinese Stock", "Market" to query, no longer repeat them here.

Search engine-based document indexing method and data search with this application Corresponding to the embodiment of the query method, the present application also provides an embodiment of a search engine based document index server and a data query server.

Referring to FIG. 5, it is a block diagram of a first embodiment of a search engine-based document indexing server of the present application:

The document index server includes an acquisition unit 510, a word segmentation unit 520, a determination unit 530, and an index unit 540.

The obtaining unit 510 is configured to obtain a document to be indexed, and the word segmentation unit 520 is configured to perform a word segmentation operation on the document obtained by the obtaining unit 510 to obtain a unary word segmentation; the determining unit 530 is configured to determine whether each of the unary segmentation words is filtered. a wording unit 540, configured to, when the determining unit 530 determines that the unary word segmentation is a filter word, the first-part word segment and the at least one unary word segment adjacent to the unitary word segmentation sequence into a multi-word segmentation, The multi-partition word is indexed, and when the judging unit 530 judges that the unary participle is not a filter word, the index is directly indexed.

6 is a block diagram of a second embodiment of a search engine-based document index server according to the present application: the document index server includes: a preset unit 610, a loading unit 620, an obtaining unit 630, a word segmentation unit 640, The judging unit 650 and the index unit 660.

The preset unit 610 is configured to preset a filter word list, and the loading unit 620 is configured to load the filter word list in the threshold unit 610; The obtaining unit 630 is configured to obtain a document to be indexed; the word segmentation unit 640 is configured to perform a word segmentation operation on the document obtained by the obtaining unit 630 to obtain a unary word segmentation; and the determining unit 650 is configured to determine whether each of the unary segmentation words is a filter word; The indexing unit 660 is configured to, when the determining unit 650 determines that the unary segmentation word is a filter word, the first-part word segment and the at least one unary word segment adjacent to the unitary word segmentation sequence into a multi-word segmentation, and the multi-partition word segmentation The index is established. When the determining unit 650 determines that the unary participle is not a filter word, the index is directly indexed. Preferably, the indexing unit 660 indexes the binary participle and the unary participle adjacent to the unitary participle into a binary participle, and indexes the binary participle.

Specifically, the determining unit 650 may include (not shown in FIG. 6): a traversal word segment unit for traversing the unary word segmentation; and a search filter word unit for determining whether each unary segmentation word is determined by searching the filter word list. Filter words.

The embodiment of the search engine based data query server provided by the present application performs data query by applying an index established by the embodiment of the document index server.

FIG. 7 is a block diagram of a first embodiment of a search engine-based data query server according to the present application. The data query server includes: an obtaining unit 710, a word segmentation unit 720, a determining unit 730, a searching unit 740, and a merging unit. 750.

The obtaining unit 710 is configured to obtain the data to be queried; The word segmentation unit 720 is configured to perform a word segmentation operation on the data acquired by the obtaining unit to obtain a unary word segmentation; the determining unit 730 is configured to determine whether each of the unary segmentation words is a filter word; the searching unit 740 is configured to determine by the determining unit 730 When the unary word segmentation is a filter word, the unary segmentation word and at least one unary word segment adjacent to the unitary word segmentation sequence are multivariate word segmentation, and the index is searched according to the multivariate word segment, when the judging unit 730 judges When the unary participle is not a filter word, the index is searched according to the unary participle; the merging unit 750 is configured to merge the query results obtained by the searching unit 740 after searching the index.

FIG. 8 is a block diagram of a second embodiment of a search engine-based data query server according to the present application. The data query server includes: a loading unit 810, an obtaining unit 820, a word segmentation unit 830, a determining unit 840, and a search. Unit 850 and merging unit 860.

The loading unit 810 is configured to load a pre-set filter word list; the obtaining unit 820 is configured to obtain the data to be queried; and the word segmentation unit 830 is configured to perform a word segmentation operation on the data acquired by the obtaining unit 820 to obtain a unary word segmentation; The determining unit 840 is configured to determine whether each of the unary word segments is a filter word; the searching unit 850 is configured to: when the determining unit 840 determines that the unary segmentation word is a filter word, the unitary word segmentation and the unary word segmentation are The at least one unary word segment adjacent to the sequence is a multi-part word segment, and the index is searched according to the multi-word segmentation. When the determining unit 840 determines that the unary segmentation word is not a filter word, the index is searched according to the unary segmentation word; Preferably, the searching unit 850 divides the unary segmentation and the unary segmentation adjacent to the unary segmentation into a binary segmentation, and searches the index according to the binary segmentation; the merging unit 860 is configured to use the searching unit The query results obtained after the 850 finds the index are merged.

Specifically, the determining unit 840 may include (not shown in FIG. 8): a traversal word segment unit for traversing the unary word segmentation; and a search filter word unit for determining whether each unary segmentation word is determined by searching the filter word list. Filter words.

Through the description of the above embodiments, the embodiment of the present application converts the query of the high frequency word into the low frequency word by dividing the high frequency unit into the finite multivariate word segment in the indexing and query process of the search engine. The query reduces the load of the search engine based on the correct query result and improves the query performance of the search engine.

It will be apparent to those skilled in the art from the above description of the embodiments that the present application can be implemented by means of a software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product in essence or in the form of a software product, which may be stored in a storage medium such as a ROM/RAM or a magnetic disk. , CD, etc., including a number of instructions to make a computer device (can be a personal computer, server, or network The methods described in the various embodiments of the present application or in certain portions of the embodiments are performed.

The various embodiments in the present specification are described in a gradual manner, and the same or similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.

This application can be used in many general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, small computers, large Computer, decentralized computing environment including any of the above systems or devices, and so on.

The application can be described in the general context of computer-executable instructions executed by a computer, such as a program module. Generally, a program module includes routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types. The present application can also be practiced in a decentralized computing environment in which tasks are performed by remote processing devices that are coupled through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

While the present invention has been described by way of example, it will be understood by those skilled in the art Do not violate the spirit of this application.

510‧‧‧Acquisition unit

520‧‧‧ segmentation unit

530‧‧‧judging unit

540‧‧‧ index unit

610‧‧‧Preset unit

620‧‧‧Loading unit

630‧‧‧Acquisition unit

640‧‧‧word segmentation unit

650‧‧‧judging unit

660‧‧‧ index unit

710‧‧‧Acquisition unit

720‧‧‧word segmentation unit

730‧‧‧judging unit

740‧‧‧Search unit

750‧‧‧Merge unit

810‧‧‧Loading unit

820‧‧‧Acquisition unit

830‧‧‧ segmentation unit

840‧‧‧judging unit

850‧‧‧Search unit

860‧‧‧Merge unit

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings to be used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a few embodiments described in the present application, and other drawings can be obtained from those skilled in the art without any inventive labor.

1 is a flow chart of a first embodiment of a search engine-based document indexing method according to the present application; FIG. 2 is a flow chart of a second embodiment of a search engine-based document indexing method according to the present application; The flow chart of the first embodiment of the search engine based data query method; FIG. 4 is a flow chart of the second embodiment of the search engine based data query method of the present application; FIG. 5 is a search engine based document of the present application; FIG. 6 is a block diagram of a second embodiment of a search engine-based document index server according to the present application; FIG. 7 is a search engine-based data query server of the present application. First embodiment block diagram; FIG. 8 is a search engine based data query server of the present application A block diagram of the second embodiment.

Claims (14)

A document indexing method based on a search engine, comprising: obtaining a document to be indexed, and performing a word segmentation operation on the document to obtain a unary word segmentation; determining whether each unary segmentation word is a filter word, wherein the filter word is included in the query During the period, the search engine ignores the high frequency word that is not checked. If the unary participle is a filter word, the unary participle and the at least one unary participle adjacent to the unary participle are multivariately segmented, and the multivariate participle is indexed. If the unary participle is not a filter word, the index of the unary participle is directly indexed. The method of claim 1, further comprising: presetting a list of filter words. The method of claim 2, wherein before the obtaining the document to be indexed, the method further comprises: loading the filter word list. The method of claim 1, wherein the unifying participle and the at least one unary participle adjacent to the unary participle are multivariately included: the unitary participle and the unitary participle are adjacent to the unitary participle The unary participle is a binary participle. The method of claim 4, wherein the unifying participle and the unary participle adjacent to the unary participle are binary words: when the unary participle is the first unary participle in the document When, The unary participle and its subsequent unary participle become a binary participle; when the unary participle is the last unary participle in the document, the unary participle is combined with its previous unary participle into a binary participle; when the unary participle is not in the document When the first unary participle and the last unary participle are combined, the unary participle is combined with the previous unary participle and the latter unary participle to form a binary participle. A search engine-based data query method, characterized in that the method uses an index established by the document indexing method described in claim 1 of the patent application, including: obtaining data to be queried, and performing word segmentation operation on the data a unitary word segmentation; determining whether each unitary word segmentation is a filter word, wherein the filter word includes a high frequency word that is ignored by the search engine during the query, and if the unitary word segmentation is a filter word, the unitary word segmentation and The at least one unary word segment adjacent to the unitary word segmentation is into a multivariate word segmentation, and the index is searched according to the multivariate word segmentation. If the unary segmentation word segment is not a filter word, the index is searched according to the unary segmentation word; the query result obtained by searching the index is performed. merge. The method of claim 6, wherein before the obtaining the data to be queried, the method further comprises: loading a preset filter word list. The method of claim 6, wherein the unary participle and the at least one unary participle adjacent to the unary participle are multivariately included: the unary participle and the order of the unary participle The unary participle is a binary participle. The method of claim 8, wherein the unifying participle and the unary participle adjacent to the unary participle are binary words: when the unary participle is the first unary participle in the material When the unitary participle is formed into a binary participle with the latter unitary participle; when the unitary participle is not the first unary participle of the material, the unitary participle is merged with the previous unary participle or the latter unary participle into a binary participle . A document indexing server based on a search engine, comprising: an obtaining unit, configured to obtain a document to be indexed; a word segmentation unit, configured to perform a word segmentation operation on the document obtained by the obtaining unit to obtain a unitary word segmentation; For determining whether each unary participle is a filter word, wherein the filter word includes a high frequency word that is ignored by the search engine during the query; and an index unit for determining, when the judging unit determines that the unary participle is filtering In the case of a word, the unitary participle and the at least one unary participle adjacent to the unitary participle sequence are multivariate word segmentation, and the multivariate participle is indexed. When the judging unit judges that the unary participle is not a filter word, the unitary participle is directly Indexing. The server according to claim 10, wherein the method further comprises: Preset unit for presetting the list of filter words. The server of claim 11, further comprising: a loading unit, configured to load the filter word list before the obtaining unit acquires the document to be indexed. A search engine-based data query server, wherein the server applies an index established by a document index server according to claim 10 of the patent application scope, comprising: an obtaining unit, configured to acquire data to be queried; a word segmentation unit, configured to perform a word segmentation operation on the data obtained by the obtaining unit to obtain a unary word segmentation; a determining unit, configured to determine whether each of the unary segmentation words is a filter word, wherein the filter word is included in the query engine and is ignored by the search engine a high frequency word that is not checked; a search unit configured to, when the determining unit determines that the unary participle is a filter word, the multipart word segment and the at least one unary word segment adjacent to the unary word segmentation into a multivariate word segment, according to the multivariate word segmentation The index is searched. When the judging unit judges that the unary participle is not a filter word, the index is searched according to the unary participle; and the merging unit is used to merge the query results obtained by the search unit after searching the index. The server of claim 13, further comprising: a loading unit, configured to load a preset filter word list before the obtaining unit acquires the data to be queried.