CN112860840A - Search processing method, device, equipment and storage medium - Google Patents

Abstract

The present application discloses a search processing method, apparatus, device and storage medium, and relates to the technical field of intelligent search. The specific implementation scheme is: obtaining the query word input by the user; determining at least one candidate keyword matching the query word based on a pre-established index relationship; after clustering each keyword according to semantics, the index relationship is obtained for the clustering Some keywords in the keywords corresponding to each category, establish the index relationship between the word segmentation of some keywords and the corresponding keywords; determine the correlation between each candidate keyword and the query word, and filter out the correlation to meet the set conditions The candidate keywords are selected; based on the selected candidate keywords, the target keywords that match the query words are determined. The embodiment of the present application reduces the search time, improves the search efficiency, and improves the recall capability of the search system through the introduction and use of the index relationship between the word segmentation and the corresponding keywords for some keywords of each category after clustering.

Description

A search processing method, apparatus, device and storage medium

technical field

The present application relates to computer technology, in particular to the technical field of intelligent search.

Background technique

After the search engine receives the query word input by the user, it needs to determine the keyword matching the query word according to the index table, and display corresponding information (such as an advertisement corresponding to the keyword) to the user according to the matching result.

In the prior art, when determining the keywords matching the query word according to the index table, the correlation between the query word and each candidate keyword in the index table needs to be determined respectively through a correlation check model, and the keywords are determined according to the correlation degree. match.

However, the correlation verification model takes a long time, and can only calculate the correlation between a small number of candidate keywords and query words each time, which greatly limits the recall ability of the search system.

SUMMARY OF THE INVENTION

The embodiments of the present application provide a search processing method, apparatus, device, and storage medium, so as to reduce the search time, improve the search efficiency, and at the same time improve the recall capability of the search system.

In a first aspect, an embodiment of the present application provides a search processing method, including:

Get the query words entered by the user;

Based on a pre-established index relationship, at least one candidate keyword matching the query term is determined; wherein, the index relationship is the corresponding classification for each category obtained by clustering each keyword according to semantics Some keywords in the keywords, the established index relationship between the participles of some keywords and the corresponding keywords;

determining the degree of relevancy between each of the candidate keywords and the query word, and screening out the candidate keywords whose degree of relevancy satisfies a set condition;

Based on the selected candidate keywords, a target keyword matching the query word is determined.

In the embodiment of the present application, the query words input by the user are obtained; based on the pre-clustering of the keywords according to the sentences, for some keywords in the keywords corresponding to the categories obtained by the clustering, the established part of the keywords The index relationship between the segmented words and the corresponding keywords is used to determine at least one candidate keyword that matches the query word; The selected candidate keywords are determined to determine the target keywords that match the query words. The above technical solution establishes an index relationship between the word segmentation of some keywords and the corresponding keywords for keywords with the same semantics, rather than establishing an index relationship for all keywords, so the candidate key determined based on the index relationship that matches the query word The number of words will also be reduced, and the workload of calculating the correlation between candidate keywords and query words will also be reduced, thereby reducing search time, improving search efficiency, and improving the recall capability of the search system.

Optionally, the method for establishing the index relationship includes:

Cluster each keyword according to semantics;

For each category obtained after clustering, select a keyword in the current category as the representative meta keyword;

A word segmentation process is performed on the representative meta-keyword, and an index relationship between each segmented word obtained by the word segmentation process and the corresponding representative meta-keyword is established.

In an optional embodiment of the above application, the keywords are clustered according to semantics, and for each category obtained after the clustering, a keyword in the current category is selected as the representative meta keyword, and the representative meta keyword is selected. The keyword is processed by word segmentation, and the index relationship between each word segment obtained by the word segmentation process and the corresponding representative meta-keyword is established, so as to improve the establishment mechanism of the index relationship and lay the foundation for the matching operation of query words. When the established index relationship is used to match the query word and determine the correlation between the matched candidate keywords and the query word, the calculation amount of the query word matching duration and the correlation is reduced, so as to reduce the search duration and improve the search efficiency.

Optionally, after selecting a keyword in the current classification as the representative meta-keyword, the method further includes:

Establish the index relationship between each representative meta-keyword and the corresponding classification;

Correspondingly, based on the selected candidate keywords, determine the target keywords matching the query words, including:

For all or part of the selected candidate keywords, determine the current candidate keyword as the classification corresponding to the representative meta keyword;

Each keyword in the determined category is read, and each read keyword is used as a target keyword matching the query word.

In an optional embodiment of the above application, in the process of constructing the index relationship, after the keywords are clustered according to semantics, the index relationship between the representative meta-keyword and a category is established, so that the target key In the process of word determination, through an index relationship corresponding to the word segmentation and the representative meta-keyword, and another index relationship corresponding to the representative meta-keyword and the classification, the determination of the target keyword is realized, the coverage of the target keyword is improved, and the coverage of the target keyword is guaranteed. Comprehensiveness of target keywords.

Optionally, for some of the selected candidate keywords, determine the current candidate keyword as the classification corresponding to the representative meta-keyword, including:

Determine a preset number of candidate keywords with the highest correlation with the query word among the selected candidate keywords;

For each determined candidate keyword, the current candidate keyword is determined as the category corresponding to the representative meta keyword.

In an optional embodiment of the above application, when the classification corresponding to the representative meta-keyword is determined, the candidate keywords are selected based on the degree of relevancy and the preset number, and the target keyword is only based on the candidate keywords with higher degree of relevancy. The key word is determined, thereby improving the matching degree between the target key word and the query word.

Optionally, select a keyword in the current category as the representative meta keyword, including:

Select a keyword with the shortest length in the current category as the representative meta-keyword.

In an optional embodiment of the above application, the process of determining the representative meta-keyword is refined to select a keyword with the shortest length in the current classification as the representative meta-keyword, thereby ensuring that the representative meta-keyword is the same as that in the current classification. The representativeness of other keywords provides a guarantee for the correlation between query words and candidate keywords.

Optionally, cluster each keyword according to semantics, including:

Mining synonymous keyword pairs for each keyword, and determining at least one keyword group based on the mined keyword pairs;

The semantics of the keywords in each keyword group are all the same.

In an optional embodiment of the above application, when the keywords are clustered, synonymous keyword pairs are mined, and keyword groups are determined based on the mined keyword pairs, which improves the clustering of keywords. Class mechanism, and at the same time, through the method of synonym mining, the comprehensiveness of the classification results is guaranteed.

Optionally, after determining the target keyword matching the query term, the method further includes:

Based on the target keyword, retrieve the information to be delivered of the information delivery party who purchased the target keyword;

Push the to-be-delivered information to the user's client for display.

In an optional embodiment of the above application, after the target keyword is determined, the to-be-posted information of the information releaser of the target keyword is determined, and the to-be-posted information is pushed to the user client for display, thereby realizing user-based query operations. The function of information push is improved, and the matching degree and comprehensiveness between the pushed information and the user's query words are improved.

In a second aspect, an embodiment of the present application further provides a search processing device, including:

The query word acquisition module is used to obtain the query word input by the user;

The candidate keyword matching module is used to determine at least one candidate keyword matching the query word based on the pre-established index relationship; Part of the keywords in the keywords corresponding to each category obtained by the class, and the established index relationship between the word segmentation of the part of the keywords and the corresponding keywords;

A candidate keyword screening module, configured to determine the degree of relevancy between each candidate keyword and the query word, and filter out the candidate keywords whose degree of relevancy satisfies a set condition;

A target keyword determination module, configured to determine a target keyword matching the query word based on the selected candidate keywords.

In a third aspect, an embodiment of the present application also provides an electronic device, including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method as provided by an embodiment of the first aspect Search for processing methods.

In a fourth aspect, embodiments of the present application further provide a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to cause the computer to perform a search process as provided by the embodiments of the first aspect method.

Other effects of the above-mentioned optional manners will be described below with reference to specific embodiments.

Description of drawings

The accompanying drawings are used for better understanding of the present solution, and do not constitute a limitation to the present application. in:

1 is a flowchart of a search processing method in Embodiment 1 of the present application;

2 is a flowchart of a search processing method in Embodiment 2 of the present application;

3 is a flowchart of a search processing method in Embodiment 3 of the present application;

4 is a structural diagram of a search processing apparatus in Embodiment 4 of the present application;

FIG. 5 is a block diagram of an electronic device used to implement the search processing method of the embodiment of the present application.

Detailed ways

Exemplary embodiments of the present application are described below with reference to the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted from the following description for clarity and conciseness.

Example 1

FIG. 1 is a flowchart of a search processing method in Embodiment 1 of the present application. The embodiment of the present application is applicable to a situation where a user searches for information through a search engine, and the method is executed by a search processing device that uses software and/or It is implemented in hardware and is specifically configured in an electronic device with a certain data computing capability.

A search processing method as shown in Figure 1, including:

S101. Obtain a query word input by a user.

The query word may be a noun or a verb input by the user and other real words.

S102. Determine at least one candidate keyword matching the query word based on a pre-established index relationship; wherein, the index relationship is a result of clustering each keyword according to semantics, for each classification obtained by the clustering. For some keywords in the corresponding keywords, the established index relationship between the participles of the partial keywords and the corresponding keywords.

The index relationship may be an inverted index.

The candidate keyword may be a content word or a phrase including at least two content words.

The index relationship may be pre-stored locally on the electronic device, in other storage devices associated with the electronic device, or in the cloud; correspondingly, when at least one candidate keyword matching the query word needs to be determined based on the index relationship, the local, Obtain or query the index relationship in other storage devices or the cloud associated with the electronic device.

S103: Determine the degree of correlation between each of the candidate keywords and the query word, and filter out the candidate keywords whose degree of correlation satisfies a set condition.

Wherein, the correlation between the candidate keyword and the query word may be determined by mutual information (MI, Mutual Information). Of course, the correlation may also be determined by other parameters in the prior art, which will not be repeated here.

The correlation between each candidate keyword found and the query word is determined respectively, and the candidate keywords are screened according to the size of the correlation and the number of candidate keywords.

Exemplarily, the relevancy of each candidate keyword may be sorted, and a set number of candidate keywords with the highest relevancy between the candidate keyword and the query word are screened out. Wherein, the set number can be set by technical personnel according to needs or experience values. Of course, the set number may be a preset number of candidate keywords to be screened, or a preset percentage of candidate keywords to be screened, and according to the preset percentage, determine the number of candidate keywords to be screened The number of candidate keywords.

S104. Based on the selected candidate keywords, determine a target keyword matching the query word.

Exemplarily, based on the selected candidate keywords, the target keywords matching the query words are determined, and the selected candidate keywords may be directly used as the target keywords matching the query words; or optional , it is also possible to determine at least one keyword in the category to which the candidate keyword belongs as the target keyword according to the category to which the candidate keyword belongs in the index relationship; or alternatively, it can also be based on the category to which the candidate keyword belongs in the index relationship. , determine the associated category of the category to which the candidate keyword belongs, and use at least one keyword in the associated category as the target keyword. The associated categories may be categories with similar semantics, similar uses, or the same upper-level category. Exemplarily, when the determined candidate keyword is "flower pot", the associated category may be "flower vase", "flower stand" or "flower fertilizer", etc.

In the embodiment of the present application, the query words input by the user are obtained; based on the index relationship between the word segmentation of the keywords and the corresponding keywords established for some keywords in each category after clustering the keywords according to the sentences in advance , determine at least one candidate keyword that matches the query word; determine the correlation between each candidate keyword and the query word, and screen out candidate keywords whose correlation meets the set conditions; matching target keywords. The above technical solution establishes an index relationship between the word segmentation of some keywords and the corresponding keywords for keywords with the same semantics, rather than establishing an index relationship for all keywords, so the candidate key determined based on the index relationship that matches the query word The number of words will also be reduced, and the workload of calculating the correlation between candidate keywords and query words will also be reduced, thereby reducing search time, improving search efficiency, and improving the recall capability of the search system.

On the basis of the technical solutions of the above-mentioned embodiments, in order to apply the search processing method to the application scenario of information push, after determining the target keyword matching the query word, based on the target keyword, search and purchase the target keyword. The information to be delivered by the information delivery party; the information to be delivered is pushed to the user's client for display.

It is understandable that the search engine usually recommends information according to the content searched by the user, so as to increase the information delivery efficiency of the search engine. After the target keywords are determined based on the query words entered and exited by the user, the information delivery party of each target keyword will be determined based on the target keywords, and then the information to be delivered corresponding to the current period will be determined according to the information delivery party; the determined information to be delivered will be pushed. To the user's client or display the web page, to display the information, so as to realize the information push to the user according to the user's query words, thereby improving the matching degree of the pushed information and the user's needs.

Embodiment 2

FIG. 2 is a flowchart of a search processing method in Embodiment 2 of the present application. The embodiments of the present application are optimized and improved on the basis of the technical solutions of the foregoing embodiments.

Further, before executing "determine at least one candidate keyword matching the query word based on the pre-established index relationship", the establishment operation of the index relationship is added; further, the establishment operation of the index relationship is refined as " Cluster each keyword according to semantics; for each category obtained after clustering, select a keyword in the current category as a representative meta-keyword; perform word segmentation processing on the representative meta-keyword, and establish a word segmentation process to obtain The index relationship between each participle and the corresponding representative meta-keyword”, to improve the establishment mechanism of the index relationship, thus laying the foundation for the determination of candidate keywords.

A search processing method as shown in Figure 2, including:

S201. Cluster each keyword according to semantics.

Exemplarily, by training and using a machine learning model, each keyword can be semantically classified, and each keyword can be divided into different categories.

Exemplarily, synonymous keyword pairs may also be mined for each keyword, and at least one keyword group may be determined based on the mined keyword pairs; wherein, the semantics of the keywords in each keyword group are all the same, that is, each keyword group has the same semantics. Each keyword group corresponds to a classification category.

It can be understood that, since the synonymous relationship satisfies reflexivity, symmetry and transitivity, during the keyword pair mining, at least one of the reflexivity, symmetry and transitivity of the keyword can also be used to determine the Associated keywords corresponding to the keywords, and a keyword group is constructed according to the keywords and the associated keywords corresponding to the keywords.

For example, when displaying information to the user, usually when the user interacts with the message corresponding to keyword A, it also generates corresponding interactive behavior for the message corresponding to keyword B, then it is determined that keyword A and keyword B are mutually exclusive. The associated keywords can form a keyword pair; wherein, the interactive behavior can be at least one of browsing, clicking, bookmarking, placing an order, and commenting.

For another example, when the information corresponding to keyword A can be obtained by rewriting the information corresponding to keyword B, or the information corresponding to keyword A and the information corresponding to keyword B can be generated based on the same information generation template, then keyword A is determined. and keyword B are related keywords to each other, and a keyword pair can be formed.

For another example, if the information corresponding to keyword A and the information corresponding to keyword B are displayed in the same delivery area and/or the delivery area is purchased by the same information delivery party, then it is determined that keyword A and keyword B are mutually related keywords, and may Form keyword pairs.

Of course, it can be understood that if keyword A and keyword B are mutually related keywords, and keyword B and keyword C are mutually related keywords, then it can be determined that keyword A and keyword C are also mutually related keywords .

S202. For each category obtained after clustering, select a keyword in the current category as a representative meta keyword.

Exemplarily, for each category obtained after clustering, any keyword in the current category may be randomly selected as the representative meta keyword. It can be understood that, since a keyword can be a short sentence containing multiple words, in order to ensure the matching degree and relevance between the selected keyword and other keywords in the same category; typically, the current keyword can be selected. A keyword with the shortest length in the category is used as the representative meta-keyword.

For example, in a classification category, including the following keywords: "double eyelid surgery cost", "double eyelid surgery price", and "how much does eyelid surgery cost", then "double eyelid surgery cost" can be selected as the representative element Key words.

S203: Perform word segmentation on the representative meta-keyword, and establish an index relationship between each segmented word obtained by the word segmentation process and the corresponding representative meta-keyword.

S204. Obtain the query word input by the user.

S205. Determine at least one candidate keyword matching the query word based on a pre-established index relationship.

Specifically, based on the index relationship between the segmented word and the representative meta-keyword, at least one representative meta-keyword matching the query term is determined as a candidate keyword.

It can be understood that, in the process of establishing the index relationship, only one representative meta-keyword is selected from each category to establish the subsequent index relationship according to the clustering results of each keyword, so the establishment of the index relationship can be greatly reduced. It can significantly improve the efficiency of index relationship establishment.

Correspondingly, in the subsequent process of using the established index relationship to determine the candidate keywords that match the query words, the amount of data operation and the matching time for matching the candidate keywords will be correspondingly reduced, so the candidate keywords can be significantly improved. Determine efficiency. At the same time, since the number of representative meta keywords in the index relationship is significantly reduced compared to the number of keywords in the index relationship in the prior art, the corresponding number of candidate keywords determined when the candidate keyword matching is determined is also significantly reduced. .

S206: Determine the degree of relevancy between each of the candidate keywords and the query word, and filter out the candidate keywords whose degree of relevancy satisfies a set condition.

It can be understood that since the number of the determined candidate keywords is significantly reduced, the amount of data calculation and the determination time when determining the correlation between each candidate keyword and the query word will inevitably be reduced, thereby improving the efficiency of determining the correlation. .

S207. Based on the selected candidate keywords, determine a target keyword matching the query word.

It should be noted that, the index relationship establishment operations corresponding to S201 to S203 may be performed before S204, or may be performed after S204, and it is only necessary to ensure that the execution sequence is before S205.

In this embodiment of the present application, an operation of establishing an index relationship is added before at least one candidate keyword matching the query word is determined based on a pre-established index relationship, and specifically, the keywords are clustered according to their semantics. For each category obtained, select a keyword in the current category as a representative meta-keyword, and perform word segmentation processing on the representative meta-keyword, and establish an index relationship between each word segment obtained by the segmentation process and the corresponding representative meta-keyword. , thus laying the foundation for the determination of candidate keywords. At the same time, in the process of establishing the index relationship, a keyword is selected from each category as the representative meta-keyword to establish the index relationship, which greatly reduces the time cost of establishing the index relationship and reduces the subsequent candidate based on the index relationship. When the matching of keywords and the calculation of the relevance degree, time cost and calculation cost can be saved, thereby improving the search efficiency.

Embodiment 3

FIG. 3 is a flowchart of a search processing method in Embodiment 3 of the present application. On the basis of the technical solutions of the foregoing embodiments, the embodiments of the present application are optimized and improved.

Further, after the operation "cluster each keyword according to semantics", "establish an index relationship between each representative meta-keyword and the corresponding classification" is added; correspondingly, the operation "based on the selected Candidate keywords, determine the target keyword matching the query word" is refined into "For all or part of the selected candidate keywords, determine the current candidate keyword as the classification corresponding to the representative meta keyword; read the determined keyword. Each keyword in the classification, and each read keyword is used as the target keyword matching the query word", so as to improve the coverage and comprehensiveness of the target keyword.

A search processing method as shown in Figure 3, including:

S301. Cluster each keyword according to semantics.

S302. For each category obtained after clustering, select a keyword in the current category as a representative meta keyword.

S303. Perform word segmentation on the representative meta-keyword, and establish a first index relationship between each segmented word obtained by the word segmentation process and the corresponding representative meta-keyword.

S304. Establish a second index relationship between each representative meta-keyword and the corresponding category.

It should be noted that, both S303 and S304 may be executed sequentially, or may be executed simultaneously, and the embodiment of the present application does not make any limitation on the specific execution order of the two.

S305. Obtain the query word input by the user.

S306. Based on the first index relationship, determine at least one representative meta-keyword matching the query term as a candidate keyword.

Based on the pre-established first index relationship between the segmented word and the representative meta-keyword, the representative meta-keyword matching the query term is determined, and the matched representative meta-keyword is used as a candidate keyword.

S307: Determine the degree of correlation between each of the candidate keywords and the query word, and filter out the candidate keywords whose degree of correlation satisfies a set condition.

It should be noted that, during the correlation calculation process, each candidate keyword will represent each keyword in the category to which the candidate keyword belongs to perform correlation calculation. Therefore, calculating the correlation between a candidate keyword and a query word is equivalent to At the same time, the correlation between other keywords in the category to which the candidate keyword belongs and the query word is calculated, which significantly reduces the amount of data calculation in the correlation calculation process, thereby improving the recall ability of the search system.

S308. For all or part of the filtered candidate keywords, determine the classification corresponding to the current candidate keyword based on the second index relationship.

In an optional implementation manner of the embodiment of the present application, each of the selected candidate keywords may be determined as the categories corresponding to the representative meta keywords by directly using the second index relationship between the representative meta keywords and the categories.

In order to reduce the amount of data computation for determining the classification process and the number of target keywords subsequently determined based on the classification results, in another optional implementation of this embodiment of the present application, only some of the selected candidate keywords may be selected. Based on the second index relationship, the current candidate keyword is determined as the category corresponding to the representative meta-keyword. Wherein, some candidate keywords may be determined from the selected candidate keywords by at least one of random screening, the degree of correlation with the query word, and the length of the candidate keywords.

Exemplarily, for some of the selected candidate keywords, determining the classification corresponding to the current candidate keyword as the representative meta-keyword based on the second index relationship may be: determining the relationship between the selected candidate keywords and the query word. The preset number of candidate keywords with the highest degree of correlation; for each determined candidate keyword, the current candidate keyword is determined based on the second index relationship as the classification corresponding to the representative meta keyword.

S309: Read each keyword in the determined category, and use each read keyword as a target keyword matching the query word.

It can be understood that by determining and screening the candidate keywords corresponding to the query words through the first index relationship corresponding to the word segmentation and the representative meta keywords, the amount of data calculation and the time consuming in the process of determining and screening the candidate keywords can be reduced, and further. Improve search efficiency; determine the target keyword by classifying the second index relationship corresponding to the representative meta-keyword. Based on the category to which the selected candidate keyword belongs, the target keyword can be effectively covered in the category, and the target keyword can be improved. word coverage and comprehensiveness.

In the embodiment of the present application, the index relationship corresponding to the classification and the representative meta-keyword is introduced, and when the target keyword is determined, the classification to which the selected candidate keyword belongs is determined based on the index relationship, so that each category in the determined classification is determined. The keyword as the target keyword improves the coverage and comprehensiveness of the target keyword, and at the same time improves the recall ability of the search system.

Embodiment 4

FIG. 4 is a structural diagram of a search processing apparatus in Embodiment 4 of the present application. The embodiment of the present application is applicable to a situation where a user searches for information through a search engine. The apparatus is implemented by software and/or hardware, and is specifically configured to have In electronic equipment with certain data computing capability.

A search processing device 400 as shown in FIG. 4 includes: a query word acquisition module 401, a candidate keyword matching module 402, a candidate keyword screening module 403 and a target keyword determination module 404; wherein,

A query word obtaining module 401, configured to obtain a query word input by a user;

The candidate keyword matching module 402 is configured to determine at least one candidate keyword matching the query word based on the pre-established index relationship; Part of the keywords in the keywords corresponding to each classification obtained by clustering, and the established index relationship between the participles of the partial keywords and the corresponding keywords;

A candidate keyword screening module 403, configured to determine the degree of relevancy between each candidate keyword and the query word, and filter out the candidate keywords whose relevancy satisfies a set condition;

The target keyword determination module 404 is configured to determine a target keyword matching the query word based on the selected candidate keywords.

In this embodiment of the present application, the query word input by the user is obtained through the query word acquisition module; the candidate keyword matching module is used to obtain the established keywords for some keywords in each category after clustering the keywords according to the sentences in advance. The index relationship between the segmented words and the corresponding keywords is used to determine at least one candidate keyword that matches the query word; the candidate keyword screening module is used to determine the correlation between each candidate keyword and the query word, and filter out the correlation degree that satisfies the set conditions. Candidate keywords; the target keyword determination module determines the target keyword matching the query word based on the selected candidate keywords. The above technical solution establishes an index relationship between the word segmentation of some keywords and the corresponding keywords for keywords with the same semantics, instead of establishing an index relationship for all keywords, so the number of candidate keywords matching the query word is also determined. will reduce, and then the workload of calculating the correlation between candidate keywords and query words will also be reduced, thereby reducing the search time, improving the search efficiency, and at the same time, improving the recall ability of the search system.

Further, the device also includes an index relationship establishment module, including:

Clustering unit, used to cluster each keyword according to semantics;

The representative meta keyword selection unit is used to select a keyword in the current category as the representative meta keyword for each category obtained after clustering;

The first index relationship establishing unit is configured to perform a word segmentation process on the representative meta-keyword, and establish an index relationship between each segmented word obtained by the word segmentation process and the corresponding representative meta-keyword.

Further, the device also includes a second index relationship establishing unit for:

After selecting a keyword in the current category as the representative meta-keyword, establish an index relationship between each representative meta-keyword and the corresponding category;

Correspondingly, the target keyword determination module 404 includes:

a classification determination unit, configured to determine the current candidate keyword as the classification corresponding to the representative meta-keyword for all or part of the selected candidate keywords;

A target keyword determination unit, configured to read each keyword in the determined category, and use each read keyword as a target keyword matching the query word.

Further, the classification determination unit is specifically used for:

Determine a preset number of candidate keywords with the highest correlation with the query word among the selected candidate keywords;

For each determined candidate keyword, the current candidate keyword is determined as the category corresponding to the representative meta keyword.

Further, the representative meta-keyword selection unit, when performing selecting a keyword in the current classification as the representative meta-keyword, is specifically used for:

Select a keyword with the shortest length in the current category as the representative meta-keyword.

Further, the clustering unit is specifically used for:

Mining synonymous keyword pairs for each keyword, and determining at least one keyword group based on the mined keyword pairs;

The semantics of the keywords in each keyword group are all the same.

Further, the device also includes an information push module, which is specifically used for:

After determining the target keyword matching the query term, based on the target keyword, retrieve the information to be delivered of the information delivery party who purchased the target keyword;

Push the to-be-delivered information to the user's client for display.

The above-mentioned search processing apparatus can execute the search processing method provided by any embodiment of the present application, and has corresponding functional modules and beneficial effects for executing the search processing method.

Embodiment 5

According to the embodiments of the present application, the present application further provides an electronic device and a readable storage medium.

As shown in FIG. 5 , it is a block diagram of an electronic device implementing the search processing method according to the embodiment of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are by way of example only, and are not intended to limit implementations of the application described and/or claimed herein.

As shown in FIG. 5, the electronic device includes: one or more processors 501, a memory 502, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. The various components are interconnected using different buses and may be mounted on a common motherboard or otherwise as desired. The processor may process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In other embodiments, multiple processors and/or multiple buses may be used with multiple memories and multiple memories, if desired. Likewise, multiple electronic devices may be connected, each providing some of the necessary operations (eg, as a server array, a group of blade servers, or a multiprocessor system). A processor 501 is taken as an example in FIG. 5 .

The memory 502 is the non-transitory computer-readable storage medium provided by the present application. Wherein, the memory stores instructions executable by at least one processor, so that the at least one processor executes the search processing method provided by the present application. The non-transitory computer-readable storage medium of the present application stores computer instructions, and the computer instructions are used to cause the computer to execute the search processing method provided by the present application.

As a non-transitory computer-readable storage medium, the memory 502 can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the search processing method in the embodiments of the present application (for example, The search processing device 400 shown in FIG. 4 includes a query word acquisition module 401, a candidate keyword matching module 402, a candidate keyword screening module 403 and a target keyword determination module 404). The processor 501 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions and modules stored in the memory 502, that is, the method for implementing the search processing in the above method embodiments.

The memory 502 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the use of the electronic device implementing the search processing method, etc. . Additionally, memory 502 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 502 may optionally include memory located remotely relative to the processor 501, and these remote memories may be connected via a network to the electronic device performing the search processing method. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

The electronic device executing the search processing method may further include: an input device 503 and an output device 504 . The processor 501 , the memory 502 , the input device 503 and the output device 504 may be connected by a bus or in other ways, and the connection by a bus is taken as an example in FIG. 5 .

The input device 503 can receive input numerical or character information, and generate key signal input related to user settings and function control of the electronic device executing the search processing method, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick , one or more mouse buttons, trackballs, joysticks and other input devices. Output devices 504 may include display devices, auxiliary lighting devices (eg, LEDs), haptic feedback devices (eg, vibration motors), and the like. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some implementations, the display device may be a touch screen.

Various implementations of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpretable on a programmable system including at least one programmable processor that The processor, which may be a special purpose or general-purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device an output device.

These computational programs (also referred to as programs, software, software applications, or codes) include machine instructions for programmable processors, and may be implemented using high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages calculation program. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or apparatus for providing machine instructions and/or data to a programmable processor ( For example, magnetic disks, optical disks, memories, programmable logic devices (PLDs), including machine-readable media that receive machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.

To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device (eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user ); and a keyboard and pointing device (eg, a mouse or trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including acoustic input, voice input, or tactile input) to receive input from the user.

The systems and techniques described herein may be implemented on a computing system that includes back-end components (eg, as a data server), or a computing system that includes middleware components (eg, an application server), or a computing system that includes front-end components (eg, a user's computer having a graphical user interface or web browser through which a user may interact with implementations of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system may be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

A computer system can include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.

According to the technical solutions of the embodiments of the present application, the query words input by the user are obtained; after clustering the keywords according to the sentences in advance, for some keywords in the keywords corresponding to the categories obtained by the clustering, the establishment of The index relationship between the word segmentation of some keywords and the corresponding keywords, determine at least one candidate keyword that matches the query word; determine the correlation between each candidate keyword and the query word, and filter out the candidates whose correlation degree satisfies the set conditions. Keywords; based on the selected candidate keywords, determine the target keywords that match the query words. The above technical solution establishes an index relationship between the word segmentation of some keywords and the corresponding keywords for keywords with the same semantics, instead of establishing an index relationship for all keywords, so the number of candidate keywords matching the query word is also determined. will reduce, and then the workload of calculating the correlation between candidate keywords and query words will also be reduced, thereby reducing the search time, improving the search efficiency, and at the same time, improving the recall ability of the search system.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, the steps described in the present application can be performed in parallel, sequentially or in different orders, and as long as the desired results of the technical solutions disclosed in the present application can be achieved, no limitation is imposed herein.

The above-mentioned specific embodiments do not constitute a limitation on the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims (10)

1. a search processing method, is characterized in that, comprises: Get the query words entered by the user; Based on a pre-established index relationship, at least one candidate keyword matching the query term is determined; wherein, the index relationship is the corresponding classification for each category obtained by clustering each keyword according to semantics Some keywords in the keywords, the established index relationship between the participles of some keywords and the corresponding keywords; determining the degree of relevancy between each of the candidate keywords and the query word, and screening out the candidate keywords whose degree of relevancy satisfies a set condition; Based on the selected candidate keywords, a target keyword matching the query word is determined. 2. The method according to claim 1, wherein the method for establishing the index relationship comprises: Cluster each keyword according to semantics; For each category obtained after clustering, select a keyword in the current category as the representative meta keyword; A word segmentation process is performed on the representative meta-keyword, and an index relationship between each segmented word obtained by the word segmentation process and the corresponding representative meta-keyword is established. 3. method according to claim 2, is characterized in that, after selecting a keyword in current classification as representative meta-keyword, described method also comprises: Establish the index relationship between each representative meta-keyword and the corresponding classification; Correspondingly, based on the selected candidate keywords, determine the target keywords matching the query words, including: For all or part of the selected candidate keywords, determine the current candidate keyword as the classification corresponding to the representative meta keyword; Each keyword in the determined category is read, and each read keyword is used as a target keyword matching the query word. 4. The method according to claim 3, wherein, for some of the selected candidate keywords, determining the current candidate keywords as the classification corresponding to the representative meta keywords, comprising: Determine a preset number of candidate keywords with the highest correlation with the query word among the selected candidate keywords; For each determined candidate keyword, the current candidate keyword is determined as the category corresponding to the representative meta keyword. 5. method according to claim 2, is characterized in that, selects a keyword in current classification as representative meta-keyword, comprising: Select a keyword with the shortest length in the current category as the representative meta-keyword. 6. The method according to any one of claims 2-5, wherein each keyword is clustered according to semantics, comprising: Mining synonymous keyword pairs for each keyword, and determining at least one keyword group based on the mined keyword pairs; The semantics of the keywords in each keyword group are all the same. 7. The method according to any one of claims 1-5, wherein after determining the target keyword matching the query word, the method further comprises: Based on the target keyword, retrieve the information to be delivered of the information delivery party who purchased the target keyword; Push the to-be-delivered information to the user's client for display. 8. A search processing device, comprising: The query word acquisition module is used to obtain the query word input by the user; The candidate keyword matching module is used to determine at least one candidate keyword matching the query word based on the pre-established index relationship; Part of the keywords in the keywords corresponding to each category obtained by the class, and the established index relationship between the word segmentation of the part of the keywords and the corresponding keywords; A candidate keyword screening module, configured to determine the degree of relevancy between each candidate keyword and the query word, and filter out the candidate keywords whose degree of relevancy satisfies a set condition; A target keyword determination module, configured to determine a target keyword matching the query word based on the selected candidate keywords. 9. An electronic device, characterized in that, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the execution of any of claims 1-7 a search processing method. 10. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute a search processing method according to any one of claims 1-7.

Images