CN112749328B

CN112749328B - Searching method, searching device and computer equipment

Info

Publication number: CN112749328B
Application number: CN202010317625.6A
Authority: CN
Inventors: 姚创沐
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-04-21
Filing date: 2020-04-21
Publication date: 2024-01-05
Anticipated expiration: 2040-04-21
Also published as: CN112749328A

Abstract

The application relates to a search method, a search device, computer equipment and a storage medium. The method comprises the following steps: displaying a query page; an input area is displayed in the query page; when the search word is monitored to be input in the input area, aggregate prompt contents corresponding to the input search word are displayed in the query page in real time; the aggregation prompt content comprises a plurality of index categories and category content which is related to the search word under each index category; the plurality of index categories are index categories that are related to query intent of the search term among index categories supported by a search engine. The method can improve the information quantity expressed during searching.

Description

Searching method, searching device and computer equipment

Technical Field

The present disclosure relates to the field of artificial intelligence and search query technologies, and in particular, to a search method, apparatus, and computer device.

Background

With the rapid development of science and technology, the internet platform has become a mainstream platform for information query in the current society. The Internet platform has massive data, and the search engine is convenient for users to find the content meeting the demands of the users in the massive data more quickly.

When a user performs search query, a traditional search engine platform prompts a plurality of related words according to keywords included in the search words so as to help the user to quickly input words to be queried. However, in the conventional method, the auxiliary prompt is limited to the words input by the user. Therefore, the user cannot know more search related information in the process of searching and inputting, and the user has certain limitation, so that the information quantity expressed in the input process is lower.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a search method, apparatus, computer device, and storage medium capable of increasing the amount of information expressed during input.

A search method, the method comprising:

displaying a query page; an input area is displayed in the query page;

when the search word is monitored to be input in the input area, aggregate prompt contents corresponding to the input search word are displayed in the query page in real time;

the aggregation prompt content comprises a plurality of index categories and category content which is related to the search word under each index category; the plurality of index categories are index categories that are related to query intent of the search term among index categories supported by a search engine.

A search apparatus, the apparatus comprising:

the display module is used for displaying the query page; an input area is displayed in the query page;

the monitoring module is used for monitoring the search words input in the input area;

the display module is further used for displaying aggregation prompt contents corresponding to the input search words in real time in the query page when the monitoring module monitors that the search words are input in the input area;

In one embodiment, the input area is a search box. And the display module is also used for displaying a drop-down frame under the search frame when the search word is input in the search frame, and displaying the category content related to the search word under the plurality of index categories in sequence in the drop-down frame.

In one embodiment, the display module is further configured to display the category content in the drop-down frame in order of high-to-low matching degree between the category content and the user portrait, and display a corresponding index category corresponding to the category content; the user portrait refers to a user portrait of the user who inputs the search term.

In one embodiment, the display module is further configured to jump to a page corresponding to the triggered category content when the monitoring module monitors a trigger operation on the category content displayed in the aggregated prompt content; and the page which is jumped to accords with the index category to which the triggered category content belongs.

In one embodiment, the aggregated hint content further includes intent-related hint words that are related to the search term. The display module is also used for jumping to a first search result page when the monitoring module monitors the triggering operation aiming at the intention association prompt word; and displaying the searched search results by taking the triggered intention association prompt word as a search condition in the first search result page.

In one embodiment, the aggregated hint content further includes a search result review portal. The display module is further used for jumping to a second search result page when the monitoring module monitors the triggering operation of the search result checking entrance; and displaying the searched search results taking the search word as a search condition in the second search result page.

A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:

displaying a query page; an input area is displayed in the query page;

A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

displaying a query page; an input area is displayed in the query page;

The searching method, the searching device, the computer equipment and the storage medium display the index categories which are supported by a search engine and are relevant to the query intention of the search words and the aggregate prompt content of category contents relevant to the search words under each index category in real time when the search words input in the input area in the query page are monitored. That is, in the process of search input, not only the content related to the search word but also the index category supported by the search engine can be prompted, so that the user can know the information related to the search engine, and therefore, compared with the method of only carrying out auxiliary prompt on the input word, more search related information can be provided, and the expressed information quantity is improved.

A search method, the method comprising:

receiving the transmitted search term input in the query page;

screening a plurality of index categories related to query intent of a search term from index categories supported by a search engine;

acquiring category contents which are under the screened index categories and are related to the search word;

generating an aggregation prompt content corresponding to the search word according to the screened index category and the category content, and returning the aggregation prompt content; and the aggregation prompt content is used for indicating that the aggregation prompt content is displayed in the query page.

A search apparatus, the apparatus comprising:

the receiving module is used for receiving the transmitted search term input in the query page;

an index category screening module for screening a plurality of index categories related to query intent of the search term from index categories supported by the search engine;

the acquisition module is used for acquiring category contents which are under the screened multiple index categories and are related to the search word;

the aggregation prompt module is used for generating and returning aggregation prompt contents corresponding to the search words according to the screened index categories and the category contents; and the aggregation prompt content is used for indicating that the aggregation prompt content is displayed in the query page.

In one embodiment, the obtaining module is further configured to recall candidate category content related to the search term under the plurality of index categories screened; when the candidate category contents under the same index category are multiple, the candidate category contents with the preset quantity are screened from the multiple candidate category contents under the same index category according to the sequence of the text relativity between the candidate category contents and the search word from high to low, and the candidate category contents are used as category contents for generating the aggregation prompt contents.

In one embodiment, the obtaining module is further configured to determine recall modes corresponding to the screened index categories respectively according to the search term; and recalling category contents related to the search word under the index category according to a recall mode corresponding to each screened index category.

In one embodiment, the aggregation prompt module is further configured to sort the index categories according to a sequence from high to low of intent strength of the category corresponding to the screened index category, or sort the category contents according to a sequence from high to low of relevance between the category contents and the search term; and generating and returning aggregation prompt contents corresponding to the search words according to the sorted index categories and the category contents under the index categories.

In one embodiment, the aggregation hint module is further configured to obtain a user representation of a user that entered the search term; determining a degree of matching between the category content and the user representation; sorting the category contents according to the sequence of the matching degree from high to low; and generating and returning aggregation prompt contents corresponding to the search words according to the sorted index categories and the category contents under the index categories.

receiving the transmitted search term input in the query page;

acquiring category contents which are under the screened multiple index categories and are related to the search word;

receiving the transmitted search term input in the query page;

According to the searching method, the device, the computer equipment and the storage medium, when the input search word is acquired, the index categories related to the query intention of the search word can be screened out from the index categories supported by the search engine in real time, the category contents related to the search word under each index category are acquired, and the aggregate prompt contents are generated and returned to the terminal, so that the terminal can prompt the contents related to the search word and prompt the index categories supported by the search engine in the input process, and a user can know the information related to the search engine, therefore, compared with the auxiliary prompt of the input word only, more search related information can be provided, and the expressed information quantity is improved.

Drawings

FIG. 1 is a diagram of an application environment for a search method in one embodiment;

FIG. 2 is a flow diagram of a search method in one embodiment;

FIG. 3 is a schematic diagram of an interface showing aggregated hints content in one embodiment;

FIG. 4 is a diagram of the composition of entity libraries and vocabularies in one embodiment;

FIG. 5 is a diagram of index category screening in one embodiment;

FIG. 6 is a schematic diagram of a recall scheme in one embodiment;

FIG. 7 is a flow chart of a search method according to another embodiment;

FIG. 8 is a schematic diagram of a search method in one embodiment;

FIG. 9 is a block diagram of a search apparatus in one embodiment;

FIG. 10 is a block diagram showing the structure of a search apparatus according to another embodiment;

FIG. 11 is an internal block diagram of a computer device in one embodiment;

fig. 12 is an internal structural view of a computer device in another embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

The searching method provided by the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers, and portable wearable devices, and the server 104 may be implemented by a stand-alone server or a server cluster composed of a plurality of servers.

The terminal 102 may present a query page; and an input area is displayed in the query page. Upon listening to the entry of a search term in the input field, the terminal 102 may then send the listening search term to the server 104. Server 104 may then filter a plurality of index categories related to the query intent of the search term from among the index categories supported by the search engine and recall category content related to the search term under the plurality of index categories filtered. Server 104 may generate aggregate hints corresponding to the search terms based on the filtered index categories and the category content and return to terminal 102. The terminal 102 may display the aggregate prompt content corresponding to the input search term in real time in the query page.

It should be noted that, in the search method in each embodiment of the present application, the search word and the aggregate prompt content are displayed correspondingly in real time, that is, along with the change of the input progress of the search word, the displayed aggregate prompt content also changes along with the change of the input progress of the search word.

It can be appreciated that the search method in the embodiments of the present application is equivalent to using artificial intelligence technology to automatically prompt the aggregate prompt content corresponding to the search term.

Artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision.

The artificial intelligence technology is a comprehensive subject, and relates to the technology with wide fields, namely the technology with a hardware level and the technology with a software level. Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

It can be understood that, the searching method in each embodiment of the application is equivalent to using a natural language processing technology in an artificial intelligence technology to identify the query intention of the search word, so as to display a plurality of index categories related to the query intention and the category contents of corresponding recall, thereby automatically prompting the aggregate prompting content corresponding to the search word.

Natural language processing (Nature Language processing, NLP) is an important direction in the fields of computer science and artificial intelligence. It is studying various theories and methods that enable effective communication between a person and a computer in natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Thus, the research in this field will involve natural language, i.e. language that people use daily, so it has a close relationship with the research in linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic questions and answers, knowledge graph techniques, and the like.

In one embodiment, as shown in fig. 2, a search method is provided, and the method is applied to the terminal in fig. 1 for illustration, and includes the following steps:

step 202, displaying a query page; and an input area is displayed in the query page.

The query page is a page for a user to input search terms (query) to trigger a search query.

The input area is an area for receiving an input sentence.

In one embodiment, the input area may include a search box. In other embodiments, the input area may further include a selection button, i.e., for inputting the search term only by clicking on the selection, in which case a search box dedicated for user input is not required, and the input of the search term may be implemented.

Step 204, when the search word is input in the input area, displaying the aggregation prompt content corresponding to the input search word in real time in the query page; wherein, the aggregate prompt content comprises a plurality of index categories and category content which is related to search words under each index category; the plurality of index categories are index categories that are related to query intent of the search term among the index categories supported by the search engine.

It can be understood that the aggregated hint content refers to content aggregated from a plurality of hint content pieces.

In the embodiment of the application, the aggregate prompt content comprises a plurality of index categories and category content which is related to the search word under each index category. It will be appreciated that category content under multiple index categories also belongs to multiple different hints content and, therefore, may be referred to as aggregate hints content.

It will be appreciated that the aggregated prompt content queried at different times may be different for time-efficient search terms. For example, the search term is a keyword with news timeliness, and during that time, the aggregate hint content should exhibit a most relevant news result; when the timeliness expires, other relevant results may be revealed.

The plurality of index categories presented in the aggregated hint content are ones of the index categories supported by the search engine and related to query intent of the search term.

Wherein the category of index, i.e., the category or type of query, is indexed. The index categories supported by the search engine refer to categories that enable feedback of query content when searching for a query in the search engine. I.e., content related to the search term can be queried from the index category.

In one embodiment, the indexing categories include at least one of a sub-application (i.e., applet) indexing category, a public account indexing category, a tool indexing category, a multimedia content indexing category (e.g., audio, video, articles, etc.), an information indexing category (e.g., news), an entertainment indexing category (e.g., music, games, novels, etc.), and a question and answer indexing category.

The child application is an application program which can be realized in the environment provided by the parent application. The parent application is an application program carrying the child application, providing an environment for the implementation of the child application. The parent application is a native application. A native application is an application that can run directly on an operating system. The child application can be run based on the parent application without downloading, and belongs to the lightweight application. For example, the WeChat is a parent application, and the WeChat applet is a child application running in the native parent application of the WeChat.

In one embodiment, the sub-applications may include sub-applications in various scenarios such as an instant messaging sub-application, a mobile payment sub-application, and a game sub-application, which may be referred to as sub-applications as long as it satisfies the characteristic of a lightweight application running based on a parent application without downloading installation.

The public account number refers to an application account number in a public platform, and can be used for communicating with a specific group in the public platform in a text, picture, voice and video mode. In one embodiment, the public account number may be a public service number in a social platform (e.g., a public service number in a WeChat platform), a service account number in some third party paystations, or the like.

It should be noted that, the index categories supported by the search engine may be determined according to the requirements of the search engine, and the indexable categories supported by different search engines may be different.

The indexing categories supported by a search engine are now described in connection with examples. Assuming that the search engine supports a sub-application index category, when a search word a is input, if the sub-application index category is an index category related to a query intention of the search word a, then content related to the search word a may be found from the sub-application index category, i.e., a sub-application related to the search word a is queried. For another example, the search engine supports a public account index category, and if the public account index category is an index category related to the query intent of the search word a, then the content related to the search word a may be found from the public account index category, i.e., the public account related to the search word a is queried.

Specifically, the user may input the search term in the input area, the terminal may monitor the input event in the input area in real time, and when the input event of the user inputting the search term in the input area is monitored, the monitored search term may be sent to the server. The server may then filter a plurality of index categories related to the query intent of the search term from among the index categories supported by the search engine, and recall category content related to the search term under the plurality of filtered index categories. The server can generate aggregate prompt content corresponding to the search word according to the screened index categories and the content of the corresponding recall category and return the aggregate prompt content to the terminal. The terminal can display the aggregation prompt content corresponding to the input search term in real time in the query page.

In one embodiment, the server may construct a database of different categories (e.g., entity library, news vocabulary, question-answer vocabulary, pattern vocabulary, etc.) based on the aggregate analysis in advance, then filter a plurality of index categories related to query intent of the search term from among index categories supported by the search engine based on the constructed database, and recall category content related to the search term from under the plurality of index categories filtered based on the constructed database. In other embodiments, the server may also directly perform intent recognition analysis to determine multiple index categories related to query intent of the search term without having to build a database of different categories based on the aggregate analysis in advance. The server can also mine the mapping relation between the search word and the category content according to the offline big data so as to recall the category content, or directly recall the category content by using a search engine. This is not limited thereto.

In one embodiment, the server may generate a content aggregation card and return to the terminal, with the aggregated hint content being presented in the content aggregation card. The terminal can display the content aggregation card at a preset position in the query page so as to display aggregation prompt contents corresponding to the input search word in the content aggregation card.

It should be noted that the aggregate hint content includes, but is not limited to, a plurality of index categories and category content under each index category that is related to the search term. I.e. the aggregate hint content may also include other content, thereby aggregating more content.

In one embodiment, the server may sort the recalled category content, generate an aggregate hint content according to the sorted category content and the corresponding index category, and return to the terminal.

According to the searching method, when the search words input in the input area in the query page are monitored, index categories which are supported by the search engine and related to the query intention of the search words and aggregate prompt contents of category contents related to the search words under each index category are displayed in real time. That is, in the process of search input, not only the content related to the search word but also the index category supported by the search engine can be prompted, so that the user can know the information related to the search engine, and therefore, compared with the method of only carrying out auxiliary prompt on the input word, more search related information can be provided, and the expressed information quantity is improved.

In one embodiment, the input area is a search box. In step 204, when it is monitored that a search term is input in the input area, aggregate prompt content corresponding to the input search term is displayed in real time in the query page, including: and when the input of the search word in the search box is monitored, displaying a drop-down box under the search box, and displaying category contents related to the search word under the plurality of index categories in sequence in the drop-down box.

The sequential display refers to sequential display among category contents related to search words under a plurality of index categories. It should be noted that, the sequential presentation herein refers to the arrangement order between the category contents of different index categories.

In one embodiment, the category content displayed under different index categories may be ranked according to at least one of a matching degree between the category content and the user portrait, a category intent strength corresponding to the index category to which the category content belongs, and a relevance degree between the category content and the search term.

It will be appreciated that the content of categories presented under the same index category may be a preset number. The preset number may be at least one. The category content displayed under the same index category can be a preset number of category content screened out according to the sequence from high to low of text relativity between the category content and the search word. The preset number is understandable, when the preset number is 1, the content of the category with the highest text relevance with the search word under the index category is displayed under the same index category.

It should be noted that, the category contents are all displayed in a classified manner according to the corresponding index category, that is, the category contents of the same index category are displayed corresponding to the associated index category.

In the embodiment, the aggregate prompt content is generated after the category content is sequenced, so that the accuracy and the referenceability of the prompt are improved.

In one embodiment, the sequentially presenting, in the drop-down box, category content related to the search term under the plurality of index categories includes: and in the drop-down frame, displaying the category content according to the sequence of the matching degree between the category content and the user portrait from high to low, and displaying the corresponding index category corresponding to the category content.

Wherein the user portrait is a user portrait of a user who inputs the search term. The plurality of index categories are ones of the index categories supported by the search engine and related to query intent of the search term.

In one embodiment, the user entering the user search term may be the user of the login account or the user selected for the search operation.

Specifically, the server may determine the matching degree between the category content related to the search term and the user portrait under the multiple index categories, then sort the category content according to the order of the matching degree from high to low, and correspondingly return the sorted category content and the index category to the terminal. The terminal can display the category content in the drop-down frame according to the sequence of the matching degree between the category content and the user portrait from high to low, and display the corresponding index category corresponding to the category content.

It will be appreciated that the order of arrangement between the index categories is consistent with the order of arrangement between the category content under the index category.

In one embodiment, the server can perform semantic analysis on category contents through a pre-trained matching model, extract semantic features, and then perform matching degree calculation on the extracted semantic features and the user portraits to obtain matching degrees between various category contents and the user portraits.

It will be appreciated that in this way, the order of preference for category content presented by different users for the same search term is different, thereby personalizing the presentation.

To facilitate understanding of personalized cues, an illustration is now provided. For example, aiming at the same search term, namely live broadcast, obtaining category contents including public account numbers and applets related to the live broadcast, technical articles describing the technology for realizing the live broadcast and corresponding live broadcast video pictures. If the user is a technician in the internet industry (e.g., he/she frequently browses a technical blog or frequently views technical information related to live broadcast), then, when he/she searches for "× live broadcast", the technical article describing the technology for implementing live broadcast is ranked first in the drop-down box, because the technical article is the category content with the highest matching degree with the user image. Assuming the user is a professional live host, then the applet will precede the technical article when it searches "live". If the user is a normal user who likes to watch live broadcast, the small program or live video picture can be arranged in front and the scientific and technological article can be arranged in the last when the user searches for live broadcast.

In the above embodiment, the category contents are ordered according to the matching degree between the category contents under different index categories and the user portrait, so that personalized prompt can be performed for the same search word and different users in the input process, that is, the priority orders of the category contents prompted by different users are different, and the accuracy of prompt in the input process is improved.

In one embodiment, the sequentially presenting, in the drop-down box, category content related to the search term under the plurality of index categories includes: and in the drop-down frame, displaying the plurality of index categories and the category contents under the plurality of index categories according to the order of the category intention intensity corresponding to the plurality of index categories from high to low.

Wherein, category intent strength is used to characterize the degree of correlation between the index category and the query intent of the search term.

In one embodiment, the category intent strength may be determined based on category intent. That is, the category intent strength may be the category intent itself, or calculated from the category intent.

Specifically, the server may analyze correlations between query intentions of the plurality of index categories and the search term to obtain category intent intensities corresponding to the index categories, and then sort category contents under the index categories according to an order of high category intent intensities, and correspondingly return the sorted category contents and the index categories to the terminal. The terminal may display the multiple index categories and the category contents under the multiple index categories in the drop-down frame according to the order of the category intention intensity corresponding to the multiple index categories from high to low.

It can be appreciated that in this way, the index category and the corresponding category content most relevant to the user query intention can be recommended to the user, and the recommendation accuracy is improved.

In one embodiment, the sequentially presenting, in the drop-down box, category content related to the search term under the plurality of index categories includes: and in the drop-down frame, displaying the category content according to the sequence of the correlation degree between the category content and the search word from high to low, and displaying the corresponding index category corresponding to the category content. The relevance is determined according to at least one of category intent strength of the index category corresponding to the category content, text relevance between the category content and the search term, and timeliness of the category content.

Wherein, the relevance is used for representing the relevance between the category content and the search word.

It should be noted that the relevance may be used to characterize the text relevance between category content and search terms, as well as to characterize the overall relevance between category content and search terms.

In one embodiment, the relevance may be determined according to at least one of a category intent strength of an index category to which the category content corresponds, a text relevance between the category content and the search term, and a timeliness of the category content.

In one embodiment, the server may score the text relevance between the category content and the search term to obtain a text relevance score, score the timeliness of the category content to obtain a content timeliness score, obtain the category intent score of the index category corresponding to the category content, and then weight and sum the text relevance score, the content timeliness score and the category intent score to obtain a comprehensive score as the relevance between the category content and the search term.

It will be appreciated that in the time-sensitive dimension score, the content of the category content is more time-sensitive than the content of the category content that is less time-sensitive. For example, in news content matching search terms, the latest news is more efficient than the content of news that has been outdated.

Specifically, the server may sort the content of each category according to the order of the high-low correlation, and correspondingly return the sorted category content and the index category to the terminal. And the terminal can display the category content in a drop-down frame according to the sequence of the correlation degree between the category content and the search word from high to low, and display the corresponding index category corresponding to the category content.

In one embodiment, the aggregated prompt further includes at least one of an intent-to-associate prompt related to the search term and a search result review portal. It will be appreciated that when the content aggregation card is generated, at least one of intent-to-associate prompt and search result review entries may be presented in the content aggregation card in addition to the content of the plurality of index categories and corresponding recalls.

Wherein the intent-associated prompt is a prompt related to the query intent.

In one embodiment, after the terminal sends the search term to the server, the server may search for the intention association prompt term related to the query intention in addition to filtering index categories related to the query intention of the search term from index categories supported by the search engine and performing recall processing of the content of the corresponding categories.

In one embodiment, the intent association hint may include at least one of an association key and an extension key. Associated keywords refer to terms that include keywords in the search term. That is, some of the associated keywords are keywords in the search term. The expanded keywords are keywords that are not included in the keywords among the search terms but are related to the query intent. That is, keywords in the search term are not included in the expanded keywords.

It will be appreciated that both the native keywords and the expanded keywords may be used as intent-related cues for the user when entering search terms to assist the user's search query.

In one embodiment, the server may also generate a search result review portal. A search result view portal for directing entry into a search result page to view entries for complete search results.

In one embodiment, the input area is a search box. When the search word is monitored to be input in the input area, the content aggregation card can be displayed at a preset position below the search box so as to display aggregation prompt content corresponding to the input search word in real time.

FIG. 3 is a schematic diagram of an interface showing aggregated hints in one embodiment. Referring to fig. 3, shown in dashed box 302 is a search box. The search term "WeChat public classes" is input in the search box. It will be appreciated that the content in the dashed box 304 in fig. 3 is the aggregation hint content shown in the content aggregation card. The "public number", "applet", "article" and "video" in the dashed box 306 are the index categories supported by the search engine and related to the query intent of the search term, and the content corresponding to each of them is the category content related to the search term under the index category. For example, "WeChat public classes" after the "public number" index category is recalled public account information related to the search term "WeChat public classes". That is, it is found by searching that there is a public account called "WeChat public classes". Similarly, as can be seen from fig. 3, the search finds that there are an article of "WeChat public classes +", weChat public classes PRO lectures ", and a video of" 2020 WeChat public classes ". In fig. 3, "related search" means the search result of the prompt word meeting the query intention, "WeChat public lesson PRO" and "WeChat short content" following the "related search" means the intention related prompt word. The WeChat public classes PRO is related keywords and comprises a keyword of a search word WeChat public classes. The "WeChat short content" is an expanded keyword which does not include the keyword "WeChat public classes" as a search term, but is an expanded keyword related to query intention. The "explore more search results" in fig. 3 is the search results view entry. The search result viewing entrance can be displayed at the bottom of the content aggregation card.

In the above embodiment, in addition to the multiple strongly related categories and category contents, the content of the aggregate prompt is displayed with the intention association prompt word and/or the search result viewing entry, so that the input path of the user can be shortened and the search efficiency can be improved through the intelligent prompt of the intention association prompt word. In addition, when the search result viewing entry is displayed, the search result page can be entered when the content (category content and intention related prompt word) prompted in the aggregate prompt content does not meet the user requirement, so that the complete search result is seen, and the search success rate is improved.

In one embodiment, the aggregated prompt further includes intent-related prompts related to the search term. The method further comprises the steps of: when the triggering operation of the category content displayed in the aggregation prompt content is monitored, jumping to a page corresponding to the triggered category content; and the jumped page accords with the index category to which the triggered category content belongs.

It will be appreciated that the user may trigger the presented aggregated prompt (e.g., a click-to-press trigger operation).

Specifically, the user can trigger the category content displayed in the aggregated prompt content, and when the terminal monitors the trigger operation on the category content displayed in the aggregated prompt content, the terminal jumps to a page corresponding to the triggered category content.

In one embodiment, when the terminal monitors the triggering operation on the category content displayed in the aggregated prompt content, the terminal can request the page information to be skipped from the server. The server can determine the page information corresponding to the triggered category content and feed the page information back to the terminal. And the terminal jumps to the page corresponding to the triggered category content from the query page according to the page information.

It will be appreciated that the page to which the jump is made corresponds to the index category to which the triggered category content belongs. For example, the trigger is the category content under the sub-application index category, and the page that jumps to is the page that accords with the sub-application index category. For example, in connection with fig. 3, assuming that the category "WeChat public classes +" under the applet index category is triggered, the applet page "WeChat public classes +" can be jumped to.

In one embodiment, the page corresponding to the triggered category content may include any one of a floor page and an account entry page of the triggered category content. Wherein, the landing page is also called landing page or guiding page. In the internet, a landing page is a web page that is displayed to a user after searching using a search engine. The account entry is a page for logging in the account.

In the embodiment, in the input process, the prompt content related to the query intention is prompted through the aggregation prompt content, and after the user triggers the prompted category content, the user directly jumps to the page corresponding to the triggered category content, so that the user can enter a specific page more quickly, and the user does not need to enter a search result page first and then select the specific page to enter, thereby improving the search query efficiency.

In one embodiment, the aggregated hint content further includes a search result view entry. The method further comprises the steps of: when triggering operation aiming at the intention association prompt word is monitored, jumping to a first search result page; the first search result page displays the searched search results by taking the triggered intention association prompt word as a search condition.

Specifically, the user may also trigger (e.g., click on) intent-related alert words in the aggregated alert content. When the terminal monitors triggering operation aiming at the intention association prompt word, the terminal can initiate a search request to the server by taking the intention association prompt word as a search condition, and the server can search by taking the triggered intention association prompt word as the search condition and generate a first search result page according to the search result and feed the first search result page back to the terminal. The terminal may then jump from the query page to the first search result page.

In the embodiment, through intelligent prompt of the intention association prompt word, the search query can be performed by directly selecting the intention association prompt word without inputting a complete sentence, so that the input path of a user is shortened, and the search efficiency is improved.

In one embodiment, the method further comprises: when triggering operation aiming at the search result checking entrance is monitored, jumping to a second search result page; and displaying the searched results by taking the search word as a search condition in the second search result page.

Specifically, the user may also trigger (e.g., click on) a search result view entry in the aggregated prompt content. When the terminal monitors the triggering operation aiming at the search result checking entrance, the terminal can initiate a search request to the server by taking the input search word as a search condition, and the server can search by taking the search word as the search condition and generate a second search result page according to the search result and feed the second search result page back to the terminal. The terminal may then jump from the query page to a second search results page.

In the above embodiment, when the content of the prompt does not meet the requirement, the search result checking entry may be directly triggered to check the complete search result, thereby improving the search success rate.

In one embodiment, the plurality of index categories in the aggregated prompt content are category intent analysis is performed on the search term to obtain category intent corresponding to each index category supported by the search engine, and the index categories are screened from the index categories supported by the search engine according to the category intent corresponding to each index category.

Wherein category intent score is used to characterize the degree of correlation between the query intent of the search term and the index category.

It will be appreciated that the degree of correlation between the query intent of the search term and the indexing category is positively correlated with the category intent. The higher the category intent score corresponding to an index category, the higher the degree of correlation between the query intent of the search term and the index category, whereas the lower the category intent score corresponding to an index category, the lower the degree of correlation between the query intent of the search term and the index category.

Specifically, after the terminal sends the search word to the server, the search word is used for instructing the server to analyze the category intention of the search word, so as to obtain category intention corresponding to each index category supported by the search engine, and the index category selected from the index categories supported by the search engine is selected according to the category intention corresponding to each index category. That is, index categories related to query intent of a search term are screened from index categories supported by a search engine according to category intent.

In one embodiment, after the terminal sends the search word to the server, the search word is used for instructing the server to input the search word into a pre-trained classification model, and predicting the score corresponding to the search word under each index category supported by the search engine as the category intention score corresponding to each index category supported by the search engine.

In one embodiment, the server has pre-built therein databases of different categories. At least one category content of the index category is prestored in each database, and category content under the same index category is stored in the same database. Each category of content has a corresponding index category label. It is understood that the index categories characterized by the index category labels are index categories supported by the search engine.

In one embodiment, some of the category content in the database may already have an index category tag and some of the category content may not have an index category tag. In this case, the corresponding index category label may be added to the category content without the index category label by manually adding, analyzing the user behavior log, and the pattern string matching process.

In one embodiment, the server may extract the index category clicked by the user during the history search from the user behavior log, and use the clicked index category as the index category label of the category content searched by the server.

In one embodiment, the pattern string matching process refers to extracting a fixed query pattern string from the category content to which the index category label is not added, then matching the fixed query pattern string with the category content to which the index category label is added, and using the index category label corresponding to the matched category content as the index category label of the category content to which the index category label is not added.

The server may perform machine learning training in advance according to preset category content and corresponding index category labels stored in the database, so as to train and obtain the classification model. Then, the server can input the search word sent by the terminal into the pre-trained classification model, and predicts the corresponding score of the search word under each index category supported by the search engine, namely, the category intention score corresponding to each index category supported by the search engine is obtained.

In one embodiment, the classification model may be a FastText or XGBoost based classification model.

In one embodiment, the pre-built databases of different categories may include a library of entities (i.e., a taxonomy database) and a vocabulary.

Wherein entities are things that exist objectively and are distinguishable from each other. In the case of databases, an entity often refers to a collection of something. The individual of each class of data objects is referred to as an entity. It can be understood that the entity library is a classification database constructed by aggregating and classifying the content of the preset category in advance. Category content belonging to the same index category is aggregated in the same entity library.

It can be understood that the data content in the entity library is the name.

In one embodiment, the entity library may include an account entity library and a name entity library. For example, the names of public account and applet account can respectively form two account entity libraries, and the music names, the novel names and the like can also respectively form corresponding name entity libraries. It can be understood that the aggregate classification of the entity library can be set according to actual requirements, which is not limited. It will be appreciated that each entity library may be one or more.

Taking fig. 4 as an example, in the example in fig. 4, "account library", "music library" and "game name" are names of different entity libraries (i.e. different index categories), and "WeChat sports" and "king" are names (i.e. category contents) specifically stored in the entity libraries.

The vocabulary is obtained by aggregating and classifying the content of the preset category in advance. Category content belonging to the same index category is aggregated in the same vocabulary.

In one embodiment, the vocabulary may include at least one of a pattern vocabulary, an information vocabulary, a question-answer vocabulary, a tool vocabulary, and an entry vocabulary. It will be appreciated that each vocabulary may be one or more.

In one embodiment, the server may be based on a manually preconfigured fixed query pattern string. Aggregation analysis can also be performed from preset category content to extract a fixed query pattern string. Thereby obtaining a pattern vocabulary according to the fixed query pattern string. It is appreciated that the fixed query pattern strings support regular etc. matching patterns (i.e., satisfy generic regular matching expressions). As shown in fig. 4, the "weather pattern" and the "route query" are each provided with a fixed query pattern string, so that corresponding word lists can be constructed respectively. For example, the weather pattern is a fixed query pattern string of "× weather", and when querying different cities or regions, only the contents in "× need to be replaced, so that the weather pattern string is provided with the fixed query pattern string, and thus, a corresponding vocabulary can be created for the weather pattern.

In one embodiment, the server may perform cluster analysis on the hot information title and content to obtain news topic words, or obtain manually edited news topic words, and then construct a news vocabulary according to the extracted news topic words. As shown in FIG. 4, news topic words are extracted for the hot spot information to obtain "new coronaries pneumonia".

In one embodiment, the server may extract keywords from the title and content of the question and answer result to obtain question and answer keywords, or obtain manually edited question and answer keywords, and then construct a question and answer word list according to the obtained question and answer keywords. As shown in fig. 4, the question-answering keywords are extracted from the medical questions and answers to obtain "how cold is handled".

In one embodiment, the server may aggregate the vocabulary entries in the vocabulary entry word stock to obtain the vocabulary entry word list. As shown in fig. 4, "scenic spot encyclopedia" and "character encyclopedia" are the results of the aggregate analysis, where "hometown" is the specific entry content.

In the above embodiment, the category intent analysis is performed on the search term, and the index category is selected from the index categories supported by the search engine according to the category intent corresponding to each index category. The index category meeting the query intention of the user can be screened out more accurately, and the accuracy and the relevance of the prompt in the input process are improved.

In one embodiment, the plurality of index categories in the aggregated prompt content are the previous preset number of index categories with the category intent greater than or equal to the preset intent threshold after the category intent scores corresponding to the index categories supported by the search engine are ranked.

It may be appreciated that, in this embodiment, the search term is further configured to instruct the server to rank the category intent scores corresponding to the index categories supported by the search engine, and screen the index categories with the category intent scores greater than or equal to the preset number of preset intent score thresholds. The preset number can be set according to actual needs.

Specifically, after the server performs category intent analysis on the search term to obtain category intent scores corresponding to the index categories supported by the search engine, the category intent scores corresponding to the index categories supported by the search engine may be ranked in order from high to low, and then a previous preset number of index categories with category intent scores greater than or equal to a preset intent score threshold are screened. It should be noted that, two conditions need to be satisfied simultaneously for the selected index category, that is, the category intention corresponding to the index category is greater than or equal to the preset intention threshold, and the order of the index category is preset in advance.

FIG. 5 is a diagram of index category filtering in one embodiment. Referring to fig. 5, for the search term "WeChat public classes," the search terms are ordered in order of high-to-low category intent under the index category supported by the search engine. Assuming that the preset intent score threshold is 0.5 and the first 4 index categories (i.e., the preset number is 4), then the 4 index categories of "public account", "applet", "news" and "video" may be screened out.

In the above embodiment, after the category intent scores corresponding to the index categories supported by the search engine are ranked, the number of index categories whose screened category intent scores are greater than or equal to the preset intent score threshold is selected. The index category meeting the query intention of the user can be screened out more accurately and rapidly, and the accuracy and the relevance of the prompt in the input process are improved.

In one embodiment, the category content is a hint content related to the search term under each index category selected for recall according to a recall mode corresponding to each index category selected for recall.

It may be appreciated that in this embodiment, the search term is further configured to instruct the server to determine, according to the search term, a recall mode corresponding to each of the screened index categories, and recall, according to the corresponding recall mode, category contents related to the search term under each of the screened index categories.

The server may return all the recalled category contents, or may perform advanced screening on the recalled category contents, further screen category contents with higher relevance to the search term, and then return the screened category contents to the terminal.

Specifically, after the server filters the index categories, the server may further determine recall modes corresponding to the filtered index categories respectively according to the search terms. It will be appreciated that different indexing categories may employ different recall modes, i.e., recall modes corresponding to each category, respectively, to recall the content of the corresponding category.

In one embodiment, the recall mode includes at least one of an entity library recall mode, a vocabulary recall mode, and a search engine recall mode.

And the entity library recall mode is used for indicating recall of category contents related to the search words from the entity library corresponding to the index category.

And the word list recall mode is used for indicating recall of category contents related to the search words from the word list corresponding to the index category.

And a search engine recall mode for indicating that the category content related to the search word is recalled from the search engine.

It will be appreciated that a corresponding recall pattern may be pre-set for each index category. An index category may correspond to at least one recall pattern. When one index category corresponds to a plurality of recall modes, one recall mode can be selected randomly to carry out recall processing, or the recall modes can be selected according to a preset sequence, and when the previous recall mode cannot be successfully recalled, the prompt content related to the search word under the index category is recalled according to the next recall mode.

It should be noted that the same recall mode may also include multiple sub-recall modes. There is also a priority among sub-recall modes in the same recall mode.

In one embodiment, the entity library recall mode may include both direct recall and create prefix tree recall modes. Wherein, direct recall refers to a mode of directly hitting recall from an entity library. Creating prefix tree recall means creating a prefix tree according to names in an entity library to which the index category belongs, and matching and recalling the search term and the name prefix in the prefix tree.

In one embodiment, the vocabulary recall patterns may include sub-recall patterns that match recalls with fixed query pattern strings in the pattern vocabulary, as well as sub-recall patterns that recall from specific vocabularies. It will be appreciated that recall matching a fixed query pattern string in a pattern vocabulary is equivalent to fuzzy matching recall, and can improve recall efficiency, so recall from a particular vocabulary is preferred.

In other embodiments, the recall process may also be performed by other recall means, such as offline aggregated result recall.

FIG. 6 is a schematic diagram of a recall pattern in one embodiment. Referring to FIG. 6, each index category may correspond to one or more recall patterns. For example, the index categories of public account/applet/music/game, the corresponding recall mode includes an entity library recall mode and a search engine recall mode. The entity library recall mode comprises two recall modes, namely a direct recall mode and a prefix (trie) tree recall mode. As can be seen from FIG. 6, the direct recall mode is preferred, corresponding results (i.e., category contents) are directly recalled from the entity library, if the recall cannot be directly hit, the mode of creating prefix tree recall is adopted, if the corresponding results still cannot be recalled, an inverted index can be established for the entity library names, and the recall is carried out through the search engine system, i.e., the search engine recall mode is adopted.

For another example, the tool queries the index category, and the corresponding recall mode includes a vocabulary recall mode and a search engine recall mode. The vocabulary recall mode comprises a sub recall mode matched with the fixed query pattern string in the pattern vocabulary and a sub recall mode recalled from the specific tool vocabulary. As can be seen from FIG. 6, the sub-recall mode of recall matching the fixed query pattern string in the pattern vocabulary is preferred, if the corresponding result cannot be recalled directly, the corresponding result is recalled from the tool vocabulary, if the corresponding result cannot still be recalled, the inverted index can be established for the tool category name, and recall is performed by the search engine system, namely, the recall mode of search engine is adopted.

Wherein, the inverted index takes the document as a key index. The record list item corresponding to the key word in the list records all documents with the word or word, one list item is a word list segment, and the record list item records the identification of the document and the position condition of the character in the document. It will be appreciated that creating an inverted index for a physical pool name corresponds to using the physical pool name as a key index. The reverse index is established for the tool class name, which is equivalent to taking the tool class name as a key index. Each of the indexing categories in fig. 6 is not explained one by one, and is the same as the recall principle of the indexing category already explained.

In the above embodiment, different recall modes can be adopted for different index categories, so that recall accuracy and recall efficiency are improved.

As shown in fig. 7, in one embodiment, a search method is provided, which is described by taking an example that the search method is applied to the server in fig. 1, and the method includes the following steps:

step 702, receiving the sent search term input in the query page.

Step 704, filtering a plurality of index categories related to query intent of the search term from the index categories supported by the search engine.

Step 706, obtaining category content under the filtered plurality of index categories and related to the search term.

Step 708, generating aggregate prompt contents corresponding to the search words according to the screened index categories and category contents and returning the aggregate prompt contents; and the aggregation prompt content is used for indicating that the aggregation prompt content is displayed in the query page.

In one embodiment, the aggregated hint content further includes at least one of an intent-to-associate hint word and a search result view entry that is related to the search word; the method comprises the steps that an intention association prompt word is used for jumping to a search result page obtained by searching under the condition that the intention association prompt word is used as a search condition after being triggered; and the search result checking inlet is used for jumping to a search result page obtained by searching under the condition of the search word after being triggered.

In one embodiment, filtering a plurality of index categories related to query intent of a search term from among the index categories supported by a search engine includes: performing category intention analysis on the search term to obtain category intention scores corresponding to the index categories supported by the search engine; screening the index categories supported by the search engine according to category intent corresponding to each index category; the category intent corresponding to the screened index category is higher than the category intent corresponding to the non-screened index category.

In one embodiment, performing category intent analysis on the search term to obtain category intent corresponding to each index category supported by the search engine includes: and inputting the search word into a pre-trained classification model, and predicting the corresponding score of the search word under each index category supported by the search engine as the category intention score corresponding to each index category supported by the search engine.

In one embodiment, the server may perform machine learning training in advance according to preset category content and corresponding index category labels stored in the database, so as to train to obtain the classification model.

In one embodiment, filtering the index categories supported by the search engine according to category intent corresponding to each index category includes: sorting category intent scores corresponding to the index categories supported by the search engine, and screening the index categories with the category intent scores being greater than or equal to the preset number of preset intent score thresholds.

In one embodiment, step 706 includes obtaining category content under the plurality of index categories filtered and related to the search term comprising: recall candidate category content associated with the search term under the plurality of index categories screened; when the candidate category contents under the same index category are multiple, the candidate category contents with the preset quantity are screened from the multiple candidate category contents under the same index category according to the sequence of the text relativity between the candidate category contents and the search word from high to low, and the candidate category contents are used as category contents for generating the aggregation prompt contents.

Wherein the pre-set number may be one or at least two. It should be noted that, in each embodiment of the present application, the number of "multiple" is at least two. Text relevance refers to the relevance between the text of category content and the text of search terms.

It can be understood that when the number of candidate category contents under the same index category is multiple, the candidate category contents with the preset number before are screened from the plurality of candidate category contents according to the sequence from high to low of the text relevance, and the candidate category contents are used for generating the aggregate prompt content, so that the category contents which the user wants to search can be more accurately recommended, and the recommendation accuracy is improved.

In one embodiment, generating aggregate hints corresponding to search terms based on the filtered index categories and category content and returning includes: sorting the index categories according to the order of the category intention intensity corresponding to the screened index categories from high to low; and generating aggregate prompt contents corresponding to the search words according to the sorted index categories and category contents under the index categories and returning the aggregate prompt contents.

Specifically, the server may analyze correlations between query intents of the plurality of index categories and the search term to obtain category intent intensities corresponding to the index categories, then sort category contents under the index categories according to an order of high category intent intensities, generate aggregate prompt contents corresponding to the search term according to the sorted index categories and category contents under the index categories, and return the aggregate prompt contents to the terminal.

In one embodiment, generating aggregate hints corresponding to search terms based on the filtered index categories and category content and returning includes: sorting the category contents according to the sequence of the high-low correlation degree between the category contents and the search words; and generating aggregate prompt contents corresponding to the search words according to the sorted index categories and category contents under the index categories and returning the aggregate prompt contents.

Specifically, the server may analyze a degree of correlation between the category content and the search term according to at least one of a category intent strength of the index category corresponding to the category content, a text correlation between the category content and the search term, and a timeliness of the category content. The server may order the category content in order of high-to-low relevance between the category content and the search term. The server can generate aggregate prompt contents corresponding to the search words according to the sorted index categories and category contents under the index categories, and return the generated aggregate prompt contents to the terminal.

In one embodiment, generating and returning aggregated hints corresponding to search terms based on the filtered index categories and category content, including: acquiring a user portrait of a user inputting search terms; determining a degree of matching between category content and the user representation; sorting the category contents according to the sequence of the matching degree from high to low; and generating aggregate prompt contents corresponding to the search words according to the sorted index categories and category contents under the index categories and returning the aggregate prompt contents.

In one embodiment, recalling category content related to a search term under the filtered plurality of index categories includes: determining recall modes corresponding to the screened index categories respectively according to the search words; and respectively indexing recall modes corresponding to the category aiming at each screened index category, and recalling category contents related to the search word under the index category.

In one embodiment, the server may sort the category content under each index category in order of high to low category intent for each index category. The server may also score the category content according to at least one of text relevance between the recalled category content and the search term and timeliness of the category content, and rank the category content in order of high score to low score. It will be appreciated that the server may also combine (e.g., weight sum) the scoring scores with the category intent scores, determine the composite scores, and order the category content in order of the composite scores from high to low. This is not limited thereto.

It will be appreciated that in the time-efficient dimension scoring, the time-efficient category content scores higher than the time-efficient category content. For example, for news content, the most recent news is scored higher than the news that has been outdated among news that matches the search term.

According to the searching method, the device, the computer equipment and the storage medium, when the input search word is acquired, the index categories related to the query intention of the search word and the category contents related to the recall search word under each index category can be screened out in real time, the aggregate prompt contents are generated and returned to the terminal, so that the terminal can prompt the contents related to the search word and the index categories supported by the search engine in the input process, and a user can know the information related to the search engine, therefore, compared with the auxiliary prompt of only the input word, more search related information can be provided, and the expressed information quantity is improved.

Fig. 8 is a schematic diagram of a search method in one embodiment. Referring to fig. 8, a user inputs a search term in a search box in a query page of a terminal, the terminal transmits the search term to a server, and the server performs category intention analysis on the search term according to pre-established databases of different categories (e.g., entity library, news vocabulary of news syndication, pattern vocabulary, question-answer vocabulary, etc.), and then screens index categories related to the query intention of the search term according to the analysis result. The server may determine recall ways corresponding to the screened index categories (different index categories have different recall ways), and recall the category content (i.e., recall the index category results) under the selected index category according to the determined recall ways, in combination with the established database (e.g., entity library, news vocabulary, pattern vocabulary, question-answer vocabulary, etc.). And then, the server can sort the recalled category contents, combine the recalled category contents to obtain an aggregation prompt content, generate a content aggregation card, and the terminal can display the content aggregation card in real time under a search box.

It should be understood that, although the steps in the above-described flowcharts are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described above may include a plurality of steps or stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of execution of the steps or stages is not necessarily sequential, but may be performed in turn or alternately with at least a part of other steps or stages.

In one embodiment, as shown in fig. 9, a search apparatus is provided, which may employ a software module or a hardware module, or a combination of both, as part of a computer device, and the apparatus specifically includes: a presentation module 902 and a listening module 904, wherein:

a display module 902, configured to display a query page; an input area is displayed in the query page.

And a listening module 904, configured to listen to the search term input in the input area.

The presentation module 902 is further configured to, when the listening module 904 listens to the input of the search term in the input area, present, in real time, the aggregated prompt content corresponding to the input search term in the query page.

Wherein, the aggregate prompt content comprises a plurality of index categories and category content which is related to search words under each index category; the plurality of index categories are index categories that are related to query intent of the search term among the index categories supported by the search engine.

In one embodiment, the input area is a search box. The display module 902 is further configured to display a drop-down box under the search box when it is monitored that a search word is input in the search box, and sequentially display category contents related to the search word under the plurality of index categories in the drop-down box.

In one embodiment, the display module 902 is further configured to display, in the drop-down frame, the category content in an order from high to low in a matching degree between the category content and the user portrait, and display a corresponding index category corresponding to the category content; the user portrait refers to a user portrait of the user who inputs the search term.

In one embodiment, the presentation module 902 is further configured to, when the monitoring module 904 monitors a triggering operation on the category content presented in the aggregated prompt content, jump to a page corresponding to the triggered category content; and the jumped page accords with the index category to which the triggered category content belongs.

In one embodiment, the aggregated hint content further includes at least one of an intent-to-associate hint word related to the search term and a search result view portal; the intention association prompt word is a prompt word related to the query intention.

In one embodiment, the aggregated hint content further includes intent-related hint words that are related to the search term. The display module 902 is further configured to jump to the first search result page when the monitoring module 904 monitors a trigger operation for the intent-to-associate prompt; the first search result page displays the searched search results by taking the triggered intention association prompt word as a search condition.

In one embodiment, the aggregated hint content further includes a search result review portal. The display module 902 is further configured to jump to a second search result page when the monitor module 904 monitors a trigger operation for the search result viewing portal; and displaying the searched results by taking the search word as a search condition in the second search result page.

In one embodiment, the category content is a hint content associated with the search term under each of the index categories selected for recall according to the recall mode to which each of the index categories selected corresponds.

In one embodiment, the recall mode includes at least one of an entity library recall mode, a vocabulary recall mode, and a search engine recall mode; an entity library recall mode for indicating recall of category content related to the search term from an entity library corresponding to the index category; the entity library is a classification database constructed by aggregating and classifying the content of the preset category in advance; category content belonging to the same index category is aggregated in the same entity library; a vocabulary recall mode, which is used for indicating that the category content related to the search word is recalled from the vocabulary corresponding to the index category; the vocabulary is constructed by aggregating and classifying the content of the preset category in advance; category content belonging to the same index category is aggregated in the same vocabulary.

In one embodiment, the indexing categories include at least one of a sub-application indexing category, a public account indexing category, a tool indexing category, a multimedia content indexing category, and a question-answer indexing category.

In one embodiment, as shown in fig. 10, a search apparatus is provided, which may employ a software module or a hardware module, or a combination of both, as part of a computer device, and the apparatus specifically includes: a receiving module 1002, an index category screening module 1004, an obtaining module 1006, and an aggregation hint module 1008, wherein:

the receiving module 1002 is configured to receive the sent search term input in the query page.

An index category filtering module 1004 is configured to filter a plurality of index categories related to query intent of the search term from index categories supported by the search engine.

An obtaining module 1006, configured to obtain category content under the screened multiple index categories and related to the search term.

The aggregation prompt module 1008 is configured to generate, according to the filtered index category and category content, an aggregation prompt content corresponding to the search term and return the aggregation prompt content; and the aggregation prompt content is used for indicating that the aggregation prompt content is displayed in the query page.

In one embodiment, the obtaining module 1006 is further configured to recall candidate category content related to the search term under the plurality of index categories that are screened; when the candidate category contents under the same index category are multiple, the candidate category contents with the preset quantity are screened from the multiple candidate category contents under the same index category according to the sequence of the text relativity between the candidate category contents and the search word from high to low, and the candidate category contents are used as category contents for generating the aggregation prompt contents.

In one embodiment, the obtaining module 1006 is further configured to determine recall modes corresponding to the screened index categories respectively according to the search term; and respectively indexing recall modes corresponding to the category aiming at each screened index category, and recalling category contents related to the search word under the index category.

In one embodiment, the aggregation prompt module 1008 is further configured to sort the index categories according to a sequence from high to low of the category intent strength corresponding to the screened index categories, or sort the category contents according to a sequence from high to low of the relevance between the category contents and the search term; and generating and returning aggregation prompt contents corresponding to the search words according to the sorted index categories and the category contents under the index categories.

In one embodiment, the aggregate prompt module 1008 is further configured to obtain a user representation of a user inputting the search term; determining a degree of matching between the category content and the user representation; sorting the category contents according to the sequence of the matching degree from high to low; and generating and returning aggregation prompt contents corresponding to the search words according to the sorted index categories and the category contents under the index categories.

The specific limitation of the search device may be referred to as limitation of the search method hereinabove, and will not be described herein. The various modules in the search apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 11. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a search method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 12. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing search data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a search method.

It will be appreciated by those skilled in the art that the structures shown in fig. 11 and 12 are block diagrams of only some of the structures associated with the present application and are not intended to limit the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.

In one embodiment, a computer-readable storage medium is provided, storing a computer program which, when executed by a processor, implements the steps of the method embodiments described above.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.

Claims

1. A method of searching, the method comprising:

displaying a query page; an input area is displayed in the query page; the input area comprises a search box; the query page is a page for a user to input search words so as to trigger a search engine to search for a query from an Internet platform;

when the search word is monitored to be input in the search box, a drop-down box is displayed under the search box in the query page, and in the drop-down box, category contents are sequenced and displayed according to at least one of matching degree between the category contents and user portrait, category intention strength corresponding to index categories to which the category contents belong and relativity between the category contents and the search word, and corresponding index categories are displayed corresponding to the category contents; the user portrait refers to a user portrait of a user who inputs the search term; the category content is included in the aggregate hint content; the aggregation prompt content is arranged in a content aggregation card; the aggregate prompt content is a content formed by aggregating a plurality of prompt contents;

The aggregate prompt content also comprises a plurality of index categories and intention association prompt words related to the search words; the plurality of index categories are index categories which are screened from index categories supported by a search engine and are relevant to query intention of the search word, wherein the category intention analysis is carried out on the search word by the server after the monitored search word is sent to the server; the category content is the category content related to the search word under the recalled index category after the index category is screened out; the index categories supported by the search engine refer to categories capable of feeding back query contents when searching for queries in the search engine;

when triggering operation aiming at the intention association prompt word is monitored, jumping to a first search result page; and displaying the searched search results taking the triggered intention association prompt words as search conditions in the first search result page.

2. The method of claim 1, wherein the indexing categories include at least one of a sub-application indexing category, a public account indexing category, a tool indexing category, a multimedia content indexing category, an information indexing category, an entertainment indexing category, or a question-answer indexing category.

3. The method according to claim 1, wherein the method further comprises:

when the triggering operation of the category content displayed in the aggregation prompt content is monitored, then

Jumping to a page corresponding to the triggered category content; and the page which is jumped to accords with the index category to which the triggered category content belongs.

4. The method of claim 1, wherein the intent association hint word includes at least one of an association keyword and an expansion keyword; the associated keywords refer to words comprising keywords in the search words; the expanded keywords are keywords that are not included in the keywords in the search term but are related to the query intent.

5. The method of any one of claims 1 to 4, wherein the aggregated hint content further comprises a search result view portal; the method further comprises the steps of:

when the triggering operation of the search result checking entrance is monitored, then

Jumping to a second search result page;

and displaying the searched search results taking the search word as a search condition in the second search result page.

6. A method of searching, the method comprising:

receiving the transmitted search term input in the query page; the query page is a page for a user to input search words to trigger a search engine to search for a query from an Internet platform;

performing category intent analysis on the search term to screen a plurality of index categories related to query intent of the search term from index categories supported by a search engine; the index categories supported by the search engine refer to categories capable of feeding back query contents when searching for queries in the search engine;

sorting the index categories according to the order of the intensity of category intention corresponding to the screened index categories from high to low, sorting the category contents according to the order of the relativity between the category contents and the search word from high to low, or sorting the category contents according to the order of the matching degree between the category contents and the user portrait from high to low; generating aggregate prompt contents corresponding to the search words according to the sorted index categories, the category contents under the index categories and the intention association prompt words related to the search words, and returning the aggregate prompt contents; the user portrait refers to a user portrait of a user who inputs the search term; the aggregation prompt content is used for indicating to display a drop-down frame under a search frame in the query page and displaying a content aggregation card in the drop-down frame so as to display the aggregation prompt content in the content aggregation card; the aggregate prompt content is a content formed by aggregating a plurality of prompt contents; the index categories supported by the search engine refer to categories capable of feeding back query contents when searching for queries in the search engine; and the intention association prompt word is used for jumping to a search result page obtained by searching under the condition that the intention association prompt word is used as a search condition after being triggered.

7. The method of claim 6, wherein the aggregated hint content further comprises a search result review portal;

and the search result checking inlet is used for jumping to a search result page obtained by searching under the condition that the search word is used as a search condition after being triggered.

8. The method of claim 6, wherein the obtaining category content under the plurality of index categories filtered and related to the search term comprises:

recall category content related to the search term under the plurality of index categories screened;

when the category content under the same index category is a plurality of

And screening the category contents of the preset quantity from a plurality of category contents under the same index category according to the sequence of the text correlation between the category contents and the search word from high to low, and taking the category contents as the category contents for generating the aggregation prompt contents.

9. The method of claim 6, wherein recalling category content related to the search term under the plurality of index categories filtered comprises:

determining recall modes corresponding to the screened index categories respectively according to the search words;

And for each screened index category, recall category contents related to the search word under the index category according to a recall mode corresponding to the index category.

10. A search apparatus, the apparatus comprising:

the display module is used for displaying the query page; an input area is displayed in the query page; the input area comprises a search box; the query page is a page for a user to input search terms to trigger search query;

the monitoring module is used for monitoring the search words input in the search box;

the display module is further configured to display a drop-down frame under the search box in the query page when the monitor module monitors that a search term is input in the input area, and in the drop-down frame, sort and display category contents according to at least one of a matching degree between category contents and user portraits, a category intention strength corresponding to an index category to which the category contents belong, and a relevance degree between the category contents and the search term, where the category contents correspond to the category contents; the user portrait refers to a user portrait of a user who inputs the search term; the category content is included in the aggregate hint content; the aggregation prompt content is arranged in a content aggregation card; the aggregate prompt content is a content formed by aggregating a plurality of prompt contents;

the display module is further used for jumping to a first search result page when the monitoring module monitors triggering operation aiming at the intention association prompt word; and displaying the searched search results taking the triggered intention association prompt words as search conditions in the first search result page.

11. The apparatus of claim 10, wherein the indexing categories include at least one of a sub-application indexing category, a public account indexing category, a tool indexing category, a multimedia content indexing category, an information indexing category, an entertainment indexing category, or a question-answer indexing category.

12. The apparatus of claim 10, wherein the presentation module is further configured to jump to a page corresponding to the triggered category content when the monitoring module monitors a trigger operation on the category content presented in the aggregated prompt content; and the page which is jumped to accords with the index category to which the triggered category content belongs.

13. The apparatus of claim 10, wherein the intent-to-associate cue words include at least one of associated keywords and expanded keywords; the associated keywords refer to words comprising keywords in the search words; the expanded keywords are keywords that are not included in the keywords in the search term but are related to the query intent.

14. The apparatus of any of claims 10 to 13, wherein the aggregated hint content further comprises a search result view entry; the display module is further used for jumping to a second search result page when the monitoring module monitors the triggering operation of the search result checking entrance; and displaying the searched search results taking the search word as a search condition in the second search result page.

15. A search apparatus, the apparatus comprising:

the receiving module is used for receiving the transmitted search term input in the query page; the query page is a page for a user to input search words to trigger a search engine to search for a query from an Internet platform;

an index category screening module for screening a plurality of index categories related to query intent of the search term from index categories supported by the search engine; the index categories supported by the search engine refer to categories capable of feeding back query contents when searching for queries in the search engine;

the acquisition module is used for acquiring category contents which are under the screened index categories and are related to the search word;

the aggregation prompt module is used for sequencing the index categories according to the order of the intensity of category intention corresponding to the screened index categories from high to low, and sequencing the category contents according to the order of the correlation degree between the category contents and the search word from high to low or the order of the matching degree between the category contents and the user portrait from high to low; generating aggregate prompt contents corresponding to the search words according to the sorted index categories, the category contents under the index categories and the intention association prompt words related to the search words, and returning the aggregate prompt contents; the user portrait refers to a user portrait of a user who inputs the search term; the aggregation prompt content is used for indicating to display a drop-down frame under a search frame in the query page and displaying a content aggregation card in the drop-down frame so as to display the aggregation prompt content in the content aggregation card; the aggregate prompt content is a content formed by aggregating a plurality of prompt contents; the index categories supported by the search engine refer to categories capable of feeding back query contents when searching for queries in the search engine; and the intention association prompt word is used for jumping to a search result page obtained by searching under the condition that the intention association prompt word is used as a search condition after being triggered.

16. The apparatus of claim 15, wherein the aggregate hint content further comprises a search result view entry; and the search result checking inlet is used for jumping to a search result page obtained by searching under the condition that the search word is used as a search condition after being triggered.

17. The apparatus of claim 15, wherein the retrieval module is further configured to recall category content related to the search term under the plurality of index categories screened; when the category content under the same index category is a plurality of category contents, the category contents with the preset quantity before are screened from the plurality of category contents under the same index category according to the sequence of the text relativity between the category contents and the search word from high to low, and the category contents are used for generating the aggregation prompt content.

18. The apparatus of claim 15, wherein the obtaining module is further configured to determine, according to the search term, a recall manner corresponding to each of the screened index categories; and for each screened index category, recall category contents related to the search word under the index category according to a recall mode corresponding to the index category.

19. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 9 when the computer program is executed.

20. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 9.