WO2022022002A1 - 一种信息展示方法、信息搜索方法及装置 - Google Patents

一种信息展示方法、信息搜索方法及装置 Download PDF

Info

Publication number
WO2022022002A1
WO2022022002A1 PCT/CN2021/093571 CN2021093571W WO2022022002A1 WO 2022022002 A1 WO2022022002 A1 WO 2022022002A1 CN 2021093571 W CN2021093571 W CN 2021093571W WO 2022022002 A1 WO2022022002 A1 WO 2022022002A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
attribute
search
screened
search result
Prior art date
Application number
PCT/CN2021/093571
Other languages
English (en)
French (fr)
Inventor
林雅婷
汪林杰
吴培培
胡胜男
Original Assignee
北京字节跳动网络技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京字节跳动网络技术有限公司 filed Critical 北京字节跳动网络技术有限公司
Priority to US18/006,715 priority Critical patent/US20230252103A1/en
Publication of WO2022022002A1 publication Critical patent/WO2022022002A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/957Browsing optimisation, e.g. caching or content distillation
    • G06F16/9577Optimising the visualization of content, e.g. distillation of HTML documents
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/31Indexing; Data structures therefor; Storage structures
    • G06F16/316Indexing structures
    • G06F16/319Inverted lists
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9538Presentation of query results
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/3331Query processing
    • G06F16/3332Query translation
    • G06F16/3334Selection or weighting of terms from queries, including natural language queries
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/335Filtering based on additional data, e.g. user or group profiles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/338Presentation of query results
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/34Browsing; Visualisation therefor
    • G06F16/345Summarisation for human users
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/65Clustering; Classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/955Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
    • G06F16/9562Bookmark management
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/279Recognition of textual entities
    • G06F40/289Phrasal analysis, e.g. finite state techniques or chunking
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis

Definitions

  • the present disclosure relates to the field of computer technology, and in particular, to an information display method, an information search method, and an apparatus.
  • a user When a user searches for a certain vertical content, he generally initiates a search request to the server through the client, and the server can search for the search result of the vertical content corresponding to the search request based on the search request, and return the search result to the client. .
  • search results There may be multiple search results corresponding to the user's search request.
  • the search results are sorted and displayed according to the historical click times of each search result. , in this way, the search results that match the user's search purpose may be ranked later, and the user will enter other search information to initiate a search request again because the user cannot find the search result he needs on the current page, resulting in the search less efficient.
  • the embodiments of the present disclosure provide at least an information display method, an information search method, and an apparatus.
  • an embodiment of the present disclosure provides an information display method, including:
  • Target search results include structured information
  • the plurality of attribute tags and at least one target search result under the attribute tags are displayed on the search result display page.
  • the plurality of acquired attribute tags have a mapping relationship with the plurality of target search results
  • the determining of at least one target search result under each of the attribute tags includes:
  • At least one target search result under each attribute tag is determined based on the mapping relationship.
  • each of the target search results carries at least one attribute tag
  • the determining of at least one target search result under each of the attribute tags includes:
  • the multiple target search results are clustered based on at least one attribute tag carried by the target search result, and at least one target search result under each of the attribute tags is determined.
  • each of the target search results carries at least one attribute tag, and a quality score of the target search result under each of the attribute tags, and the quality score is used to represent the target The degree of match between the search result and this attribute label.
  • the determining at least one target search result under each of the attribute tags includes:
  • For each target search result determine the target attribute label corresponding to the target search result based on the quality score of the target search result under each of the attribute labels carried;
  • the plurality of target search results are clustered to determine at least one target search result under each attribute label.
  • the plurality of attribute tags and at least one target search result under the attribute tags are displayed on the search result display page, including:
  • the method further includes:
  • a first trigger operation for the attribute tag is received, and the first attribute tag is determined based on the first trigger operation.
  • the plurality of attribute tags and at least one target search result under the attribute tags are displayed on the search result display page, including:
  • a structured information card in at least one target search result corresponding to the attribute tag is vertically displayed under the attribute tag, and the structured information card in the at least one target search result supports sliding display.
  • the search information is recipe search information
  • the attribute tag is used to indicate attribute information of the recipe.
  • the structured information in the target search result includes recipe multimedia information and recipe description information of the target recipe
  • the recipe description information includes ingredient information and production process
  • the structured information card The structured information is displayed in a preset form.
  • the method further includes:
  • the search result display page jumps to the information page of the target search result corresponding to the trigger operation, wherein the structured information in the target search result is from the target search result. Extracted from the information page.
  • an information search method including:
  • the index library includes a mapping relationship between information pages, structured information, and attribute tags
  • target search results include structured information
  • the multiple target search results and the multiple attribute tags are sent to the client.
  • the plurality of information pages that match the search information are determined based on the index library, including:
  • a plurality of information pages matching the search information are selected from the information pages to be screened.
  • the search type is determined on the search information, and based on the determination result, a plurality of information pages to be screened that match the search information are determined from an index library corresponding to the determination result.
  • search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result Based on the search keyword and at least one synonym corresponding to the search keyword, search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result.
  • the attribute label corresponding to each information page to be screened is determined according to the following method:
  • Pre-set at least one tag keyword corresponding to each attribute tag
  • the tag keyword is searched from the information page to be screened, and if any tag keyword is found, the attribute tag corresponding to the found tag keyword is used as the attribute tag of the information page to be screened.
  • the performing search type determination on the search information includes:
  • the judgment result corresponding to the search information is determined, wherein the judgment model is obtained by training based on the sample information and the search type label carried by the sample information.
  • the quality scores of the multiple information pages to be screened are determined by the following method:
  • the weighted summation of the text relevance and the text quality parameter is performed to obtain the quality score of the information page to be screened.
  • the determining the textual relevancy between the title, abstract, and text content of the information page to be screened includes:
  • the text relevance between the titles, abstracts, and body content of the information pages to be filtered is determined.
  • the text correlation between the title, abstract, and text content of the information page to be screened is determined based on the semantic vectors corresponding to the title, abstract, and text content of the information page to be screened.
  • the second semantic vector corresponding to the title of the information page to be screened is multiplied by the first semantic vector to obtain the text relevance.
  • the text quality parameters include image and text quality parameters and/or picture clarity.
  • the plurality of information pages matching the search information are screened from the information pages to be screened, including:
  • an intermediate information page is selected from the information pages to be screened;
  • a plurality of information pages matching the search information are filtered out from the intermediate information page.
  • the plurality of information pages matching the search information are screened from the information pages to be screened, including:
  • For each attribute label determine the information page to be filtered corresponding to the attribute label, and sort the information page to be filtered corresponding to the attribute label according to the corresponding quality score;
  • the information pages to be screened whose sorting results are ranked in the top N are taken as a plurality of information pages matching the search information, where N is a preset positive integer.
  • the target search result includes a structured information card displaying the structured information
  • the generating multiple target search results and multiple attribute tags corresponding to the search information according to the structured information and attribute tags corresponding to the multiple information pages including:
  • the structured information is added to the preset position of the structured information card, and the target search result corresponding to the information page is generated; the attributes of the information page
  • the label is the attribute label of the target search result corresponding to the information page.
  • an information display device including:
  • a receiving module for receiving the search information input by the user
  • a first determination module configured to acquire multiple target search results and multiple attribute labels corresponding to the search information, and determine at least one target search result under each of the attribute labels; wherein, the target search results include a structure information;
  • the display module is configured to display the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page.
  • the plurality of acquired attribute tags have a mapping relationship with the plurality of target search results
  • the first determination module when determining at least one target search result under each of the attribute tags, is used for:
  • At least one target search result under each attribute tag is determined based on the mapping relationship.
  • each of the target search results carries at least one attribute tag
  • the first determination module when determining at least one target search result under each of the attribute tags, is used for:
  • the multiple target search results are clustered based on at least one attribute tag carried by the target search result, and at least one target search result under each of the attribute tags is determined.
  • each of the target search results carries at least one attribute tag, and a quality score of the target search result under each of the attribute tags, and the quality score is used to represent the target The degree of match between the search result and this attribute label.
  • the first determination module when determining at least one target search result under each of the attribute tags, is used for:
  • For each target search result determine the target attribute label corresponding to the target search result based on the quality score of the target search result under each of the attribute labels carried;
  • the plurality of target search results are clustered to determine at least one target search result under each attribute label.
  • the display module when displaying the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page, is used to:
  • the first determining module is further configured to:
  • a first trigger operation for the attribute tag is received, and the first attribute tag is determined based on the first trigger operation.
  • the display module when displaying the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page, is used to:
  • a structured information card in at least one target search result corresponding to the attribute tag is vertically displayed under the attribute tag, and the structured information card in the at least one target search result supports sliding display.
  • the search information is recipe search information
  • the attribute tag is used to indicate attribute information of the recipe.
  • the structured information in the target search result includes recipe multimedia information and recipe description information of the target recipe
  • the recipe description information includes ingredient information and production process
  • the structured information card The structured information is displayed in a preset form.
  • the display module is also used for:
  • the search result display page jumps to the information page of the target search result corresponding to the trigger operation, wherein the structured information in the target search result is from the target search result. Extracted from the information page.
  • an information search apparatus including:
  • a second determining module configured to determine a plurality of information pages matching the search information based on an index library; wherein, the index library includes a mapping relationship between information pages, structured information, and attribute tags;
  • a generating module configured to generate a plurality of target search results and a plurality of attribute labels corresponding to the search information according to the structured information and attribute labels corresponding to the plurality of information pages; wherein, the target search results include structured information;
  • a sending module configured to send the multiple target search results and the multiple attribute tags to the client.
  • the second determination module when determining a plurality of information pages matching the search information based on the index library, is used for:
  • a plurality of information pages matching the search information are selected from the information pages to be screened.
  • the second determination module determines the search type of the search information, and based on the determination result, determines from the index library corresponding to the determination result that matches the search information.
  • the second determination module determines the search type of the search information, and based on the determination result, determines from the index library corresponding to the determination result that matches the search information.
  • search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result Based on the search keyword and at least one synonym corresponding to the search keyword, search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result.
  • the second determining module is further configured to determine the attribute label corresponding to each information page to be screened according to the following method:
  • Pre-set at least one tag keyword corresponding to each attribute tag
  • the tag keyword is searched from the information page to be screened, and if any tag keyword is found, the attribute tag corresponding to the found tag keyword is used as the attribute tag of the information page to be screened.
  • the second determination module when performing search type determination on the search information, is used for:
  • the judgment result corresponding to the search information is determined, wherein the judgment model is obtained by training based on the sample information and the search type label carried by the sample information.
  • the second determination module is configured to determine the quality scores of the multiple information pages to be screened by the following method:
  • the weighted summation of the text relevance and the text quality parameter is performed to obtain the quality score of the information page to be screened.
  • the second determining module when determining the textual relevancy between the title, abstract, and text content of the information page to be screened, is used to:
  • the text relevance between the titles, abstracts, and body content of the information pages to be filtered is determined.
  • the second determining module determines the title, abstract, and text content of the information page to be screened based on the semantic vectors corresponding to the title, abstract, and text content of the information page to be screened. Text affinity between when used for:
  • the second semantic vector corresponding to the title of the information page to be screened is multiplied by the first semantic vector to obtain the text relevance.
  • the text quality parameters include image and text quality parameters and/or picture clarity.
  • the second determination module based on the quality scores of the plurality of information pages to be screened, filters out a plurality of information pages that match the search information from the information pages to be screened when used for:
  • the intermediate information pages are selected from the information pages to be screened;
  • a plurality of information pages matching the search information are filtered out from the intermediate information page.
  • the second determination module based on the quality scores of the plurality of information pages to be screened, filters out a plurality of information pages that match the search information from the information pages to be screened when used for:
  • For each attribute label determine the information page to be filtered corresponding to the attribute label, and sort the information page to be filtered corresponding to the attribute label according to the corresponding quality score;
  • the information pages to be screened whose sorting results are ranked in the top N are taken as a plurality of information pages matching the search information, where N is a preset positive integer.
  • the target search result includes a structured information card displaying the structured information
  • the generating module when generating multiple target search results and multiple attribute tags corresponding to the search information according to the structured information and attribute tags corresponding to the multiple information pages, is used for:
  • the structured information is added to the preset position of the structured information card, and the target search result corresponding to the information page is generated; the attribute of the information page
  • the label is the attribute label of the target search result corresponding to the information page.
  • embodiments of the present disclosure further provide a computer device, including: a processor, a memory, and a bus, where the memory stores machine-readable instructions executable by the processor, and when the computer device runs, the processing A bus communicates between the processor and the memory, and when the machine-readable instructions are executed by the processor, the above-mentioned first aspect, or the steps in any possible implementation manner of the first aspect, or the above-mentioned first aspect is executed.
  • an embodiment of the present disclosure further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to execute the first aspect, or any one of the first aspect. Steps in one possible implementation manner, or perform the above-mentioned second aspect, or steps in any possible implementation manner of the second aspect.
  • the information display method, information search method and device provided by the embodiments of the present disclosure can display each target search result in the corresponding attribute according to the attribute label corresponding to the target search result after acquiring multiple target search results corresponding to the search information. label, so as to realize the classified display of target search results.
  • different attribute labels can be displayed to users intuitively. Users can select the attribute labels they are interested in to view the target search results under the attribute labels. On the one hand, it satisfies users' viewing requirements for target search results under different attribute labels, and on the other hand, it facilitates users to quickly locate target search results that they are interested in, thereby improving search efficiency.
  • FIG. 1 shows a flowchart of an information display method provided by an embodiment of the present disclosure
  • FIG. 2a shows a schematic diagram of a search result display page provided by an embodiment of the present disclosure
  • Fig. 2b shows a schematic diagram of another search result display page provided by an embodiment of the present disclosure
  • FIG. 3 shows a schematic diagram of another search result display page provided by an embodiment of the present disclosure
  • FIG. 4 shows a schematic flowchart of an information search method provided by an embodiment of the present disclosure
  • FIG. 5 shows a target search result provided by an embodiment of the present disclosure, and a schematic diagram of a method for determining an attribute label corresponding to each target search result
  • FIG. 6 shows a schematic diagram of the architecture of an information display apparatus provided by an embodiment of the present disclosure
  • FIG. 7 shows a schematic structural diagram of an information search apparatus provided by an embodiment of the present disclosure.
  • FIG. 8 shows a schematic structural diagram of a computer device 800 provided by an embodiment of the present disclosure
  • FIG. 9 shows a schematic structural diagram of a computer device 900 provided by an embodiment of the present disclosure.
  • the search results with no content when displaying the search results with no content, they are displayed in sequence.
  • this display method may not meet the user's search needs, and the user may need to initiate a search again based on other search terms.
  • the user's search request is "Kung Pao Chicken”
  • what the user is really interested in is "Kung Pao Chicken's Homemade Practice”
  • the search results of his interest may be ranked at the back, and the user cannot view the "Kung Pao Chicken” on the current page.
  • the search result corresponding to "Homemade Practice of Kung Pao Chicken the user may re-request with "Homemade Practice of Kung Pao Chicken", which leads to lower search efficiency.
  • the present disclosure provides an information display method, an information search method and an apparatus, which can display each target search result on In this way, different attribute tags can be displayed intuitively to users. Users can select the attribute tags they are interested in to view the target under the attribute tag.
  • the search result on the one hand, satisfies the user's viewing requirements for target search results under different attribute labels, and on the other hand facilitates the user to quickly locate the target search result that interests him, thereby improving the search efficiency.
  • the execution subject of the information display method provided by the embodiment of the present disclosure generally has a terminal device, and the terminal device may be a smart phone. , Tablet PC, Smart TV, Personal Computer, etc.
  • FIG. 1 is a flowchart of an information display method provided by an embodiment of the present disclosure, the method includes steps 101 to 103, wherein:
  • Step 101 Receive search information input by a user.
  • Step 102 Acquire multiple target search results and attribute tags corresponding to the search information, and determine at least one target search result under each attribute tag; wherein the target search results include structured information.
  • Step 103 Display the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page.
  • the user terminal may generate a search request based on the search information, and send the generated search request to a corresponding server, where the search request may carry the search information.
  • step 102 For step 102,
  • any one of the following methods can be used:
  • mapping relationship between the acquired multiple attribute tags and multiple target search results.
  • the mapping between multiple attribute tags and multiple target search results can be directly based on the mapping relationship between multiple attribute tags and multiple target search results. relationship, identifying at least one target search result under each attribute tag.
  • mapping relationship may be the mapping relationship between the attribute label and the identification information of each target search result, and the mapping relationship may be generated by the server after each target search result is determined by the server.
  • mapping relationship can be as shown in Table 1 below:
  • Each target search result carries an attribute tag.
  • the The plurality of target search results are clustered to determine at least one target search result under each of the attribute tags.
  • the target search results carrying the same attribute label can be divided into the same category, and after all target search results are divided, at least one target search result under each attribute label can be determined.
  • Each target search carries at least one attribute tag, and the quality score of the target search result under each attribute tag, where the quality score is used to indicate the matching degree between the target search result and the attribute tag , wherein the quality score corresponding to the target search result is determined by the server.
  • the corresponding target search result may be determined based on the quality score of the target search result under each of the attribute tags carried by the target search result. target attribute label; then, based on the target attribute label corresponding to each target search result, cluster the multiple target search results to determine at least one target search result under each attribute label.
  • the corresponding attribute label with the highest quality score among the attribute labels carried by the target search result may be used as the corresponding target search result.
  • target attribute tag when determining the target attribute label corresponding to each target search result, the corresponding attribute label with the highest quality score among the attribute labels carried by the target search result may be used as the corresponding target search result.
  • the search result display page when multiple attribute tags and at least one target search result under the attribute tags are displayed on the search result display page, multiple attribute tags can be displayed horizontally on the search result display page, and then the attribute tags can be displayed from the attribute tags.
  • the display when displaying multiple attribute labels horizontally, the display may be based on a preset display sequence of attribute labels.
  • the search information is recipe search information
  • the attribute labels are used to indicate recipe attribute information
  • Exemplary attribute tags may include homely, cooked, easy, famous, and the like.
  • the attribute tag corresponding to the first triggering operation is used as the first attribute tag, or the first attribute tag may be displayed in the order of the first attribute tag.
  • the first attribute tag After receiving the first triggering operation for the attribute tag, the attribute tag corresponding to the first triggering operation is used as the first attribute tag, or the first attribute tag may be displayed in the order of the first attribute tag. The first attribute tag.
  • Each attribute tag can correspond to at least one target search result.
  • it When displaying at least one target search result under the first attribute tag on the search result display page, it can be sorted according to the quality score corresponding to the at least one target search result, and the results can be sorted.
  • the at least one target search result is displayed side by side.
  • the multiple attribute tags and at least one target search result under the attribute tag are displayed on the search result display page
  • the multiple attribute tags may be displayed horizontally on the search result display page.
  • the structured information card in at least one target search result supports sliding display. Swipe up and down to switch and display multiple structured information cards under the same attribute label. When a user wants to view other natural search results other than the target search result, it can be determined by the user's sliding distance.
  • the structured information card displayed under the current attribute label can be switched. If the user's sliding distance on the user end is greater than or equal to the first preset distance, You can directly swipe up the search result display page to display other natural search results.
  • the structured information in the target search result may include recipe multimedia information and recipe description information of the target recipe, and the recipe description information includes ingredient information and production process;
  • the recipe evaluation information may include Ratings, popularity, number of favorites, and production time, etc.
  • the production process may include video production and production steps, and the structured information is displayed in a preset form in the structured information card.
  • the recipe multimedia information may display the recipe information in a multimedia format, for example, may include recipe pictures, and/or recipe videos, and the recipe video may be a video showing the recipe in a video format.
  • the multimedia information of the recipe can be displayed at the first preset position of the structured information card, and then displayed in the second preset position of the structured information card.
  • Recipe description information is displayed at the preset position.
  • Figure 3 includes four attribute tags of "homey, cooking, easy to cook, and famous", and the first attribute tag is "homely”, which is currently displayed on the page in Figure 3. is the search result with the highest quality score under the first attribute label, and the search result includes the production video, ingredients, and production steps (ie, the "practice” in the figure).
  • the above-mentioned multiple attribute tags can be displayed, and the target search result with the highest quality score under the attribute tag is displayed under the first attribute tag.
  • the target search result with the highest quality score under the second attribute tag corresponding to the trigger operation may be displayed in response to a trigger operation for other attribute tags except the first attribute tag.
  • the triggering operation for other attribute tags other than the first attribute tag may include single click, double click, long press, repress and so on.
  • the target search result with the highest quality score under the attribute label of “Place Meal” can be displayed.
  • target search results may also be displayed in response to a swipe operation for other target search results except the target search result with the highest quality score.
  • the search result display page may jump to the information page of the target search result corresponding to the trigger operation, wherein the target The structured information in the search results is extracted from this information page.
  • the information page of the target search result may display video, picture, text, comments and other information corresponding to the target search result.
  • the method provided in this disclosure can also be applied to other search information that can be explained in detail.
  • the search information may be origami tutorial search information.
  • the attribute labels corresponding to the origami tutorial search results may include easy to learn, tiny etc., the attribute tag is pre-set.
  • each target search result can be displayed under the corresponding attribute label according to the attribute label corresponding to the target search result, so as to realize Classified display of target search results.
  • the target search content under different attribute labels can be intuitively displayed to users. Users can view different target search results based on the selection of displayed attribute labels, which satisfies users The search requirements under the attribute label improve the search efficiency.
  • FIG. 4 is a schematic flowchart of an information search method provided by an embodiment of the present disclosure, the method can be applied to a server, and includes the following steps:
  • Step 401 Obtain search information.
  • Step 402 Determine a plurality of information pages matching the search information based on an index library; wherein, the index library includes a mapping relationship between information pages, structured information, and attribute tags.
  • Step 403 Generate multiple target search results and multiple attribute tags corresponding to the search information according to the structured information and attribute tags corresponding to the multiple information pages; wherein, the targeted search results include structured information.
  • Step 404 Send the multiple target search results and the multiple attribute tags to the client.
  • the acquiring search information may be acquiring a search request sent by a user terminal, where the search request carries the search information input by the user.
  • the index library stores multiple information pages, structured information extracted from each information page, and attribute tags corresponding to the information pages, and the relationship between the information pages and the structured information and attribute tags is
  • the mapping relationship may be a mapping relationship between information page identifiers, structured information identifiers, and attribute tags.
  • the method shown in FIG. 5 may be used, including the following steps:
  • Step 501 Perform a search type judgment on the search information, and based on the judgment result, from an index library corresponding to the judgment result, determine a plurality of information pages to be screened that match the search information; Information pages have corresponding attribute tags.
  • the search information when multiple information pages to be screened corresponding to the search information are determined based on the determination result, the search information may be subjected to word segmentation processing to determine at least one search keyword included in the search information; and then obtain at least one synonym corresponding to the search keyword; then based on the search keyword and at least one synonym corresponding to the search keyword, from the index library corresponding to the judgment result, search for the search keyword
  • the index library may be an inverted index library.
  • the identification result of the search information is a recipe information search
  • the identification result of the search information is an origami tutorial search
  • the origami tutorial inverted index database can be used to find a plurality of to-be-filtered information pages corresponding to the search information.
  • each piece of index information in the inverted index database corresponds to a unique index keyword, and the index information can be directly retrieved based on the index keyword.
  • At least one synonym corresponding to the word when searching for multiple information pages to be screened corresponding to the search information from the inverted index database corresponding to the judgment result, the search keyword and the search keyword can be calculated respectively.
  • the correlation between the relative synonym of the keyword and the index keyword corresponding to each index information When the correlation corresponding to any index information is greater than the preset correlation, the information page corresponding to the index information can be determined as the search information.
  • the matching information page to be filtered.
  • the index information stored in the inverted index library may include index keywords, and an inverted list indicating that the index keywords are included, and an exemplary storage form thereof may be shown in Table 1 below:
  • (x, y) indicates that the y-th position of the x-th information page includes an index keyword.
  • 1 can represent the title part
  • 2 can represent the abstract part
  • 3 represents the text part.
  • "Sauce pork” the index keyword is included in the title part of the first information page, the body part of the third information page, and the abstract part of the fourth information page.
  • a corresponding inverted list can be found, and then the found inverted list can be merged, and the information pages in the merged list are used as search information The corresponding information page to be filtered.
  • the inverted index library also stores structured information corresponding to a plurality of information pages, each information page has a corresponding preset attribute label, and the structured information corresponding to the information page may be the information page from the information page. Structured information extracted from .
  • index keyword "Kung Pao Chicken” there may be multiple information pages corresponding to the index keyword, and then the corresponding information page with more clicks may be selected, and the selected information page may be selected. Extract the structured information as the information page corresponding to the index keyword "Kung Pao Chicken” and the structured information corresponding to the information page.
  • the extracted structured information when extracting structured information from an information page, it can be extracted from multiple contents such as title, abstract, introduction, ingredients, steps, author information, etc.
  • the extracted structured information can be as shown in Table 2 below :
  • each structured information can be Displayed in structured information cards in a preset format.
  • the attribute label corresponding to each information page when storing in the index library, can be pre-stored, and specifically, the attribute label corresponding to each information page to be screened can be determined in advance according to the following method:
  • Pre-set at least one tag keyword corresponding to each attribute tag
  • the tag keyword is searched from the information page to be screened, and if any tag keyword is found, the attribute tag corresponding to the found tag keyword is used as the attribute tag of the information page to be screened.
  • the corresponding tag keyword can be set as family practice, family detailed steps, etc., and for the attribute tag "famous”, the corresponding tag keyword can be set as the author's name.
  • the search positions of different attribute tags in the index file are different. For example, for the attribute tag "homely”, it can be searched from the title part and the introduction part of the index file; For the attribute tag "famous”, it can be found from the author name section of the index file.
  • an attribute tag can be added to the index file manually.
  • the information page to be screened when determining the attribute label corresponding to each information page to be screened, can also be input into a pre-trained neural network, and the neural network can output the corresponding information page to be screened.
  • the attribute labels of wherein the neural network is obtained by training based on sample documents, and the sample documents have corresponding pre-labeled attribute labels.
  • index file stored in the inverted index library may have only one attribute tag, or may have multiple attribute tags.
  • a determination result corresponding to the search information may be determined based on the trained determination model and the search information, wherein the determination model is based on the sample information, and the It is obtained by training the search type labels carried by the sample information.
  • Step 502 Based on the quality scores of the plurality of information pages to be screened, screen out a plurality of information pages matching the search information from the information pages to be screened.
  • the text relevance between the titles, abstracts, and body content of the information pages to be screened may be determined first; and, obtaining The text quality parameter of the information page to be screened; then the weighted summation of the text relevance and the text quality parameter is performed to obtain the quality score of the information page to be screened.
  • the text quality parameters may include image and text quality parameters and/or picture clarity.
  • the word vectors corresponding to the title, abstract and text content of the information page to be screened may be input into the trained language processing model to obtain the semantic vectors corresponding to the title, abstract, and text content of the information page to be screened; then, based on the semantic vectors corresponding to the title, abstract, and text content of the information page to be filtered, determine the Text relevance between title, abstract, body content.
  • the language processing model may be a BERT model
  • the text relevance may be used to represent the degree of association between the title, abstract, and body of the information page to be screened.
  • the information to be screened The title of the page may have a low degree of association with the abstract and the body part.
  • the target search results generated based on such information pages to be filtered may lead to low precision, so the information pages to be filtered can be filtered by text relevance.
  • the text correlation between the title, abstract, and text content of the information page to be screened is determined.
  • the semantic vectors corresponding to the title, abstract, and body content of the information page to be screened can be multiplied, and the product is used as the text correlation between the title, abstract, and body content of the information page to be filtered.
  • the text correlation between the title, abstract, and body content of the information page to be screened is determined based on the semantic vectors corresponding to the title, abstract, and body content of the information page to be screened.
  • the semantic vectors corresponding to the abstract and the text content of the information page to be screened are weighted and summed to obtain a first semantic vector; and then the second semantic vector corresponding to the title of the information page to be screened is combined with the The first semantic vector is multiplied to obtain the text relevance.
  • the information pages to be screened are screened out that match the search information.
  • intermediate information pages can be selected from the information pages to be screened; and then based on the quality scores of the intermediate information pages and the corresponding intermediate search results
  • the attribute tag is used to filter out target search results from the intermediate search results.
  • the intermediate information page corresponding to each attribute tag may be determined according to the attribute tag corresponding to the intermediate information page, and for each attribute tag , sort the intermediate information pages under the attribute label according to the corresponding quality scores, and then use the top M intermediate information pages as the information pages matching the search information, where M is a preset value.
  • the number of information pages that match the search information filtered out under different attribute tags may be the same, for example There can be three.
  • the importance of different attribute labels may be different.
  • the number of information pages corresponding to the different attribute labels that match the search information can be preset. For each attribute When filtering information pages matching the search information, the number of corresponding attribute tags can be filtered.
  • a certain information page to be screened corresponds to multiple attribute tags, and the quality score of the information page to be screened under each attribute tag is relatively high, it can be used as the information under each attribute tag.
  • the attribute tag with the highest quality score of the information page to be screened can be used as the target attribute tag corresponding to the information page to be screened, and then the information page to be screened as the information page that matches the search information under the target attribute tag.
  • the targeted search results may include structured information cards displaying structured information.
  • step 404 For step 404,
  • the user terminal After sending the target search result and the attribute tag corresponding to the target search result to the user terminal, the user terminal can display the received target search result based on the information display method as shown in FIG. 1 .
  • the server when it sends the target search result to the client, it may also send other natural search results to the client, where the natural search result is a common search result found by the server based on the search information, and the target search result is different from Yes, the target search result is found based on the inverted index database corresponding to the target search result, and the general search result is found based on the database of all available search results.
  • the information display method, information search method and device provided by the embodiments of the present disclosure can display each target search result in the corresponding attribute according to the attribute label corresponding to the target search result after acquiring multiple target search results corresponding to the search information. label, so as to realize the classified display of the target search results.
  • different attribute labels can be displayed to the user intuitively.
  • the user can select the attribute label that he is interested in to view the target search results under the attribute label. On the one hand, it satisfies users' viewing requirements for target search results under different attribute labels, and on the other hand, it facilitates users to quickly locate target search results that they are interested in, thereby improving search efficiency.
  • the writing order of each step does not mean a strict execution order but constitutes any limitation on the implementation process, and the specific execution order of each step should be based on its function and possible Internal logic is determined.
  • the embodiment of the present disclosure also provides an information display device corresponding to the information display method. Reference may be made to the implementation of the method, and repeated descriptions will not be repeated.
  • the device includes: a receiving module 601 , a first determining module 602 , and a displaying module 603 ; wherein,
  • the first determination module 602 is configured to acquire multiple target search results and multiple attribute labels corresponding to the search information, and determine at least one target search result under each of the attribute labels; wherein the target search results include structured information;
  • the display module 603 is configured to display the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page.
  • the plurality of acquired attribute tags have a mapping relationship with the plurality of target search results
  • the first determining module 602 when determining at least one target search result under each of the attribute tags, is used for:
  • At least one target search result under each attribute tag is determined based on the mapping relationship.
  • each of the target search results carries at least one attribute tag
  • the first determining module 602 when determining at least one target search result under each of the attribute tags, is used for:
  • the multiple target search results are clustered based on at least one attribute tag carried by the target search result, and at least one target search result under each of the attribute tags is determined.
  • each of the target search results carries at least one attribute tag, and a quality score of the target search result under each of the attribute tags, and the quality score is used to represent the target The degree of match between the search result and this attribute label.
  • the first determining module 602 when determining at least one target search result under each of the attribute tags, is configured to:
  • For each target search result determine the target attribute label corresponding to the target search result based on the quality score of the target search result under each of the attribute labels carried;
  • the plurality of target search results are clustered to determine at least one target search result under each attribute label.
  • the display module 603 when displaying the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page, is used to:
  • the first determining module 602 is further configured to:
  • a first trigger operation for the attribute tag is received, and the first attribute tag is determined based on the first trigger operation.
  • the display module 603 when displaying the plurality of attribute tags and at least one target search result under the attribute tags on the search result display page, is used to:
  • a structured information card in at least one target search result corresponding to the attribute tag is vertically displayed under the attribute tag, and the structured information card in the at least one target search result supports sliding display.
  • the search information is recipe search information
  • the attribute tag is used to indicate attribute information of the recipe.
  • the structured information in the target search result includes recipe multimedia information and recipe description information of the target recipe
  • the recipe description information includes ingredient information and production process
  • the structured information card The structured information is displayed in a preset form.
  • the display module 603 is further used for:
  • the search result display page jumps to the information page of the target search result corresponding to the trigger operation, wherein the structured information in the target search result is from the target search result. Extracted from the information page.
  • the embodiments of the present disclosure also provide an information search apparatus corresponding to the information search method. Reference may be made to the implementation of the method, and repeated descriptions will not be repeated.
  • the apparatus includes: an acquisition module 701, a second determination module 702, a generation module 703, and a transmission module 704, specifically:
  • a second determining module 702 configured to determine a plurality of information pages matching the search information based on an index library; wherein the index library includes a mapping relationship between information pages, structured information, and attribute tags;
  • a generating module 703, configured to generate a plurality of target search results and a plurality of attribute labels corresponding to the search information according to the structured information and attribute labels corresponding to the plurality of information pages; wherein, the target search results include structure information;
  • the sending module 704 is configured to send the multiple target search results and the multiple attribute tags to the client.
  • the second determining module 702 when determining a plurality of information pages matching the search information based on an index library, is configured to:
  • a plurality of information pages matching the search information are selected from the information pages to be screened.
  • the second determination module 702 determines the search type of the search information, and based on the determination result, determines that the search information matches the search information from the index library corresponding to the determination result. When there are multiple information pages to be filtered, it is used to:
  • search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result Based on the search keyword and at least one synonym corresponding to the search keyword, search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result.
  • the second determining module 702 is further configured to determine the attribute label corresponding to each information page to be screened according to the following method:
  • Pre-set at least one tag keyword corresponding to each attribute tag
  • the tag keyword is searched from the information page to be screened, and if any tag keyword is found, the attribute tag corresponding to the found tag keyword is used as the attribute tag of the information page to be screened.
  • the second determining module 702 when determining the search type of the search information, is configured to:
  • the judgment result corresponding to the search information is determined, wherein the judgment model is obtained by training based on the sample information and the search type label carried by the sample information.
  • the second determining module 702 is configured to determine the quality scores of the multiple information pages to be screened by the following method:
  • the weighted summation of the text relevance and the text quality parameter is performed to obtain the quality score of the information page to be screened.
  • the second determining module 702 when determining the textual relevancy between the title, abstract, and body content of the information page to be screened, is used to:
  • the text relevance between the titles, abstracts, and body content of the information pages to be filtered is determined.
  • the second determining module 702 determines the title, abstract, and text of the information page to be screened based on the semantic vectors corresponding to the title, abstract, and text content of the information page to be screened.
  • textual relevance between content it is used to:
  • the second semantic vector corresponding to the title of the information page to be screened is multiplied by the first semantic vector to obtain the text relevance.
  • the text quality parameters include image and text quality parameters and/or picture clarity.
  • the second determination module 702 based on the quality scores of the plurality of information pages to be screened, filters out a plurality of information matching the search information from the information pages to be screened page, used to:
  • an intermediate information page is selected from the information pages to be screened;
  • a plurality of information pages matching the search information are filtered out from the intermediate information page.
  • the second determination module 702 based on the quality scores of the plurality of information pages to be screened, filters out a plurality of information matching the search information from the information pages to be screened page, used to:
  • For each attribute label determine the information page to be filtered corresponding to the attribute label, and sort the information page to be filtered corresponding to the attribute label according to the corresponding quality score;
  • the information pages to be screened whose sorting results are ranked in the top N are taken as a plurality of information pages matching the search information, where N is a preset positive integer.
  • the target search result includes a structured information card displaying the structured information
  • the generating module 703 when generating multiple target search results and multiple attribute tags corresponding to the search information according to the structured information and attribute tags corresponding to the multiple information pages, is used for:
  • the structured information is added to the preset position of the structured information card, and the target search result corresponding to the information page is generated; the attribute of the information page
  • the label is the attribute label of the target search result corresponding to the information page.
  • FIG. 8 a schematic structural diagram of a computer device 800 provided by an embodiment of the present disclosure includes a processor 801 , a memory 802 , and a bus 803 .
  • the memory 802 is used to store the execution instructions, including the memory 8021 and the external memory 8022; the memory 8021 here is also called the internal memory, and is used to temporarily store the operation data in the processor 801 and the data exchanged with the external memory 8022 such as the hard disk,
  • the processor 801 exchanges data with the external memory 8022 through the memory 8021.
  • the processor 801 communicates with the memory 802 through the bus 803, so that the processor 801 executes the following instructions:
  • Target search results include structured information
  • the plurality of attribute tags and at least one target search result under the attribute tags are displayed on the search result display page.
  • the plurality of acquired attribute tags have a mapping relationship with the plurality of target search results
  • the determining of at least one target search result under each of the attribute tags includes:
  • At least one target search result under each attribute tag is determined based on the mapping relationship.
  • each of the target search results carries at least one attribute tag
  • the determining of at least one target search result under each of the attribute tags includes:
  • the multiple target search results are clustered based on at least one attribute tag carried by the target search result, and at least one target search result under each of the attribute tags is determined.
  • each of the target search results carries at least one attribute label, and the quality score of the target search result under each of the attribute labels, the The quality score is used to indicate the matching degree between the target search result and the attribute tag.
  • the determining at least one target search result under each of the attribute tags includes:
  • For each target search result determine the target attribute label corresponding to the target search result based on the quality score of the target search result under each of the attribute labels carried;
  • the plurality of target search results are clustered to determine at least one target search result under each attribute label.
  • the plurality of attribute tags and at least one target search result under the attribute tags are displayed on the search result display page, including:
  • the method further includes:
  • a first trigger operation for the attribute tag is received, and the first attribute tag is determined based on the first trigger operation.
  • the plurality of attribute tags and at least one target search result under the attribute tags are displayed on the search result display page, including:
  • a structured information card in at least one target search result corresponding to the attribute tag is vertically displayed under the attribute tag, and the structured information card in the at least one target search result supports sliding display.
  • the search information is recipe search information
  • the attribute tag is used to indicate attribute information of the recipe.
  • the structured information in the target search result includes recipe multimedia information and recipe description information of the target recipe, and the recipe description information includes ingredient information and production process. , displaying the structured information in the structured information card in a preset form.
  • the method further includes:
  • the search result display page jumps to the information page of the target search result corresponding to the trigger operation, wherein the structured information in the target search result is from the target search result. Extracted from the information page.
  • a schematic structural diagram of a computer device 900 provided by an embodiment of the present disclosure includes a processor 901 , a memory 902 , and a bus 903 .
  • the memory 902 is used to store the execution instructions, including the memory 9021 and the external memory 9022; the memory 9021 here is also called the internal memory, and is used to temporarily store the operation data in the processor 901 and the data exchanged with the external memory 9022 such as the hard disk,
  • the processor 901 exchanges data with the external memory 9022 through the memory 9021.
  • the processor 901 communicates with the memory 902 through the bus 903, so that the processor 901 executes the following instructions:
  • the index library includes a mapping relationship between information pages, structured information, and attribute tags
  • target search results include structured information
  • the multiple target search results and the multiple attribute tags are sent to the client.
  • the plurality of information pages that match the search information are determined based on the index library, including:
  • a plurality of information pages matching the search information are selected from the information pages to be screened.
  • the search type judgment is performed on the search information, and based on the judgment result, from the index library corresponding to the judgment result, the search information is determined from the index library corresponding to the judgment result.
  • Multiple matching information pages to be filtered including:
  • search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result Based on the search keyword and at least one synonym corresponding to the search keyword, search multiple information pages to be screened corresponding to the search information from the index database corresponding to the determination result.
  • the attribute label corresponding to each information page to be screened is determined according to the following method:
  • Pre-set at least one tag keyword corresponding to each attribute tag
  • the tag keyword is searched from the information page to be screened, and if any tag keyword is found, the attribute tag corresponding to the found tag keyword is used as the attribute tag of the information page to be screened.
  • the search type determination on the search information includes:
  • the judgment result corresponding to the search information is determined, wherein the judgment model is obtained by training based on the sample information and the search type label carried by the sample information.
  • the quality scores of the plurality of information pages to be screened are determined by the following method:
  • a weighted summation of the text relevance and the text quality parameter is performed to obtain a quality score of the information page to be screened.
  • the determining the textual relevance between the title, abstract, and text content of the information page to be screened includes:
  • the text relevance between the titles, abstracts, and body content of the information pages to be filtered is determined.
  • the title, abstract, and content of the information page to be screened are determined based on the semantic vectors corresponding to the title, abstract, and text content of the information page to be screened.
  • Text relevance between body content including:
  • the second semantic vector corresponding to the title of the information page to be screened is multiplied by the first semantic vector to obtain the text relevance.
  • the text quality parameters include image and text quality parameters and/or picture definition.
  • Information pages including:
  • an intermediate information page is selected from the information pages to be screened;
  • a plurality of information pages matching the search information are filtered out from the intermediate information page.
  • Information pages including:
  • For each attribute label determine the information page to be filtered corresponding to the attribute label, and sort the information page to be filtered corresponding to the attribute label according to the corresponding quality score;
  • the information pages to be screened whose sorting results are ranked in the top N are taken as a plurality of information pages matching the search information, where N is a preset positive integer.
  • the target search result includes a structured information card displaying the structured information
  • the generating multiple target search results and multiple attribute tags corresponding to the search information according to the structured information and attribute tags corresponding to the multiple information pages including:
  • the structured information is added to the preset position of the structured information card, and the target search result corresponding to the information page is generated; the attribute of the information page
  • the label is the attribute label of the target search result corresponding to the information page.
  • Embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the information display method and the information search method described in the foregoing method embodiments are executed A step of.
  • the storage medium may be a volatile or non-volatile computer-readable storage medium.
  • the computer program products of the information display method and the information search method provided by the embodiments of the present disclosure include a computer-readable storage medium storing program codes, and the instructions included in the program codes can be used to execute the information described in the above method embodiments.
  • the steps of the display method and the information search method reference may be made to the foregoing method embodiments, which will not be repeated here.
  • Embodiments of the present disclosure also provide a computer program, which implements any one of the methods in the foregoing embodiments when the computer program is executed by a processor.
  • the computer program product can be specifically implemented by hardware, software or a combination thereof.
  • the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK), etc. Wait.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a processor-executable non-volatile computer-readable storage medium.
  • the computer software products are stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes. .

Abstract

一种信息展示方法、信息搜索方法及装置,该方法包括:接收用户输入的搜索信息(101);获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息(102);在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果(103)。

Description

一种信息展示方法、信息搜索方法及装置
相关申请的交叉引用
本申请要求于2020年07月31日提交的,申请号为202010760603.7、发明名称为“一种信息展示方法、信息搜索方法及装置”的中国专利申请的优先权,该申请的全文通过引用结合在本申请中。
技术领域
本公开涉及计算机技术领域,具体而言,涉及一种信息展示方法、信息搜索方法及装置。
背景技术
用户在进行某垂类内容搜索时,一般是通过用户端向服务器发起搜索请求,服务器可以基于该搜索请求,查找与搜索请求对应的垂类内容的搜索结果,并将该搜索结果返回给用户端。
与用户的搜索请求对应的搜索结果可能有多个,一般来说会根据各搜索结果的历史点击次数,将搜索结果进行排序展示。,这种方式下,可能会出现匹配用户搜索目的的搜索结果排序比较靠后的情况,用户会因为在当前页面查找不到自己需要的搜索结果而输入其他搜索信息再次发起搜索请求,从而导致搜索效率较低。
发明内容
本公开实施例至少提供一种信息展示方法、信息搜索方法及装置。
第一方面,本公开实施例提供了一种信息展示方法,包括:
接收用户输入的搜索信息;
获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息;
在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,获取的所述多个属性标签与所述多个目标搜索结果具有映射关系;
所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
基于所述映射关系确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,每个所述目标搜索结果携带有至少一个属性标签;
所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
基于所述目标搜索结果携带的至少一个属性标签,对所述多个目标搜索结果进行聚类,确定每个所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,每个所述目标搜索结果携带有至少一个属性标签,以及该所述目标搜索结果在各所述属性标签下的质量分数,所述质量分数用于表示所述目标搜索结果与该属性标签之间的匹配程度。
一种可能的实施方式中,所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
针对每个目标搜索结果,基于该目标搜索结果在携带的各所述属性标签下的质量分数,确定该目标搜索结果对应的目标属性标签;
基于每个目标搜索结果对应的目标属性标签,对所述多个目标搜索结果进行聚类,确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,所述在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果,包括:
在所述搜索结果展示页面横向展示所述多个属性标签;
从所述属性标签中选择第一属性标签,在所述第一属性标签下横向展示所述第一属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,所述方法还包括:
接收针对于所述属性标签的第一触发操作,基于所述第一触发操作确定所述第一属性标签。
一种可能的实施方式中,所述在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果,包括:
在所述搜索结果展示页面横向展示所述多个属性标签;
针对各所述属性标签,在该属性标签下纵向展示该属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,所述搜索信息为菜谱搜索信息,所述属性标签用于指示菜谱的属性信息。
一种可能的实施方式中,所述目标搜索结果中的结构化信息包含目标菜谱的菜谱多媒体信息和菜谱描述信息,所述菜谱描述信息中包含食材信息和制作流程,在所述结构化信息卡片中以预设形式展示所述结构化信息。
一种可能的实施方式中,所述方法还包括:
响应于对于任一目标搜索结果的第二触发操作,由所述搜索结果展示页面跳转至所 述触发操作对应的目标搜索结果的信息页面,其中,该目标搜索结果中的结构化信息从该信息页面中提取。
第二方面,本公开实施例提供了一种信息搜索方法,包括:
获取搜索信息;
基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系;
根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息;
将所述多个目标搜索结果和所述多个属性标签发送至用户端。
一种可能的实施方式中,所述基于索引库确定与所述搜索信息匹配的多个信息页面,包括:
对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面;其中,每个待筛选信息页面对应有属性标签;
基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,所述对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面,包括:
对所述搜索信息进行分词处理,确定所述搜索信息中包含的至少一个搜索关键词;
获取与所述搜索关键词相对应的至少一个同义词;
基于所述搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结果对应的索引库中,查找与所述搜索信息对应的多个待筛选信息页面。
一种可能的实施方式中,根据以下方法确定每个待筛选信息页面对应的属性标签:
预先设置每个属性标签对应的至少一个标签关键词;
从所述待筛选信息页面中查找所述标签关键词,若查找到任一标签关键词,将查找到的标签关键词对应的属性标签作为该待筛选信息页面的属性标签。
一种可能的实施方式中,所述对所述搜索信息进行搜索类型判定,包括:
基于训练的判定模型和所述搜索信息,确定所述搜索信息对应的判定结果,其中,所述判定模型是基于样本信息,以及所述样本信息携带的搜索类型标签训练得到的。
一种可能的实施方式中,通过以下方法确定所述多个待筛选信息页面的质量分数:
确定所述待筛选信息页面中的标题、摘要、正文内容之间的文本相关度;以及,获取所述待筛选信息页面的文本质量参数;
将所述文本相关度和所述文本质量参数进行加权求和,得到所述待筛选信息页面的质量分数。
一种可能的实施方式中,所述确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度,包括:
将所述待筛选信息页面的标题、摘要、正文内容分别对应的词向量输入至训练的语言处理模型,得到所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量;
基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
一种可能的实施方式中,所述基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度,包括:
将所述待筛选信息页面的摘要、正文内容分别对应的语义向量进行加权求和,得到第一语义向量;
将所述待筛选信息页面的标题对应的第二语义向量与所述第一语义向量相乘,得到文本相关度。
一种可能的实施方式中,所述文本质量参数,包括图文质量参数和/或图片清晰度。
一种可能的实施方式中,所述基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面,包括:
基于所述待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出中间信息页面;
基于所述中间信息页面的质量分数和所述中间信息页面对应的属性标签,从所述中间信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,所述基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面,包括:
针对每一个属性标签,确定对应该属性标签的待筛选信息页面,并将对应该属性标签的待筛选信息页面按照对应的质量分数进行排序;
将排序结果排在前N位的待筛选信息页面作为与所述搜索信息匹配的多个信息页面,N为预设正整数。
一种可能的实施方式中,所述目标搜索结果包括展示有所述结构化信息的结构化信息卡片;
所述根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签,包括:
基于与所述搜索信息匹配的每一个信息页面对应的结构化信息,在结构化信息卡片的预设位置处添加该结构化信息,生成该信息页面对应的目标搜索结果;所述信息页面的属性标签为该信息页面对应的目标搜索结果的属性标签。
第三方面,本公开实施例提供了一种信息展示装置,包括:
接收模块,用于接收用户输入的搜索信息;
第一确定模块,用于获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息;
展示模块,用于在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,获取的所述多个属性标签与所述多个目标搜索结果具有映射关系;
所述第一确定模块,在确定每个所述属性标签下的至少一个目标搜索结果时,用于:
基于所述映射关系确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,每个所述目标搜索结果携带有至少一个属性标签;
所述第一确定模块,在确定每个所述属性标签下的至少一个目标搜索结果时,用于:
基于所述目标搜索结果携带的至少一个属性标签,对所述多个目标搜索结果进行聚类,确定每个所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,每个所述目标搜索结果携带有至少一个属性标签,以及该所述目标搜索结果在各所述属性标签下的质量分数,所述质量分数用于表示所述目标搜索结果与该属性标签之间的匹配程度。
一种可能的实施方式中,所述第一确定模块,在确定每个所述属性标签下的至少一个目标搜索结果时,用于:
针对每个目标搜索结果,基于该目标搜索结果在携带的各所述属性标签下的质量分数,确定该目标搜索结果对应的目标属性标签;
基于每个目标搜索结果对应的目标属性标签,对所述多个目标搜索结果进行聚类,确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,所述展示模块,在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果时,用于:
在所述搜索结果展示页面横向展示所述多个属性标签;
从所述属性标签中选择第一属性标签,在所述第一属性标签下横向展示所述第一属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,所述第一确定模块,还用于:
接收针对于所述属性标签的第一触发操作,基于所述第一触发操作确定所述第一属性标签。
一种可能的实施方式中,所述展示模块,在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果时,用于:
在所述搜索结果展示页面横向展示所述多个属性标签;
针对各所述属性标签,在该属性标签下纵向展示该属性标签对应的至少一个目标搜 索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,所述搜索信息为菜谱搜索信息,所述属性标签用于指示菜谱的属性信息。
一种可能的实施方式中,所述目标搜索结果中的结构化信息包含目标菜谱的菜谱多媒体信息和菜谱描述信息,所述菜谱描述信息中包含食材信息和制作流程,在所述结构化信息卡片中以预设形式展示所述结构化信息。
一种可能的实施方式中,所述展示模块,还用于:
响应于对于任一目标搜索结果的第二触发操作,由所述搜索结果展示页面跳转至所述触发操作对应的目标搜索结果的信息页面,其中,该目标搜索结果中的结构化信息从该信息页面中提取。
第四方面,本公开实施例提供了一种信息搜索装置,包括:
获取模块,用于获取搜索信息;
第二确定模块,用于基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系;
生成模块,用于根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息;
发送模块,用于将所述多个目标搜索结果和所述多个属性标签发送至用户端。
一种可能的实施方式中,所述第二确定模块,在基于索引库确定与所述搜索信息匹配的多个信息页面时,用于:
对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面;其中,每个待筛选信息页面对应有属性标签;
基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,所述第二确定模块,在对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面时,用于:
对所述搜索信息进行分词处理,确定所述搜索信息中包含的至少一个搜索关键词;
获取与所述搜索关键词相对应的至少一个同义词;
基于所述搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结果对应的索引库中,查找与所述搜索信息对应的多个待筛选信息页面。
一种可能的实施方式中,所述第二确定模块,还用于根据以下方法确定每个待筛选信息页面对应的属性标签:
预先设置每个属性标签对应的至少一个标签关键词;
从所述待筛选信息页面中查找所述标签关键词,若查找到任一标签关键词,将查找到的标签关键词对应的属性标签作为该待筛选信息页面的属性标签。
一种可能的实施方式中,所述第二确定模块,在对所述搜索信息进行搜索类型判定时,用于:
基于训练的判定模型和所述搜索信息,确定所述搜索信息对应的判定结果,其中,所述判定模型是基于样本信息,以及所述样本信息携带的搜索类型标签训练得到的。
一种可能的实施方式中,所述第二确定模块,用于通过以下方法确定所述多个待筛选信息页面的质量分数:
确定所述待筛选信息页面中的标题、摘要、正文内容之间的文本相关度;以及,获取所述待筛选信息页面的文本质量参数;
将所述文本相关度和所述文本质量参数进行加权求和,得到所述待筛选信息页面的质量分数。
一种可能的实施方式中,所述第二确定模块,在确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,用于:
将所述待筛选信息页面的标题、摘要、正文内容分别对应的词向量输入至训练的语言处理模型,得到所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量;
基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
一种可能的实施方式中,所述第二确定模块,在基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,用于:
将所述待筛选信息页面的摘要、正文内容分别对应的语义向量进行加权求和,得到第一语义向量;
将所述待筛选信息页面的标题对应的第二语义向量与所述第一语义向量相乘,得到文本相关度。
一种可能的实施方式中,所述文本质量参数,包括图文质量参数和/或图片清晰度。
一种可能的实施方式中,所述第二确定模块,在基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面时,用于:
基于所述待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出中间信息页面;
基于所述中间信息页面的质量分数和所述中间信息页面对应的属性标签,从所述中间信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,所述第二确定模块,在基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面时,用于:
针对每一个属性标签,确定对应该属性标签的待筛选信息页面,并将对应该属性标签的待筛选信息页面按照对应的质量分数进行排序;
将排序结果排在前N位的待筛选信息页面作为与所述搜索信息匹配的多个信息页面,N为预设正整数。
一种可能的实施方式中,所述目标搜索结果包括展示有所述结构化信息的结构化信息卡片;
所述生成模块,在根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签时,用于:
基于与所述搜索信息匹配的每一个信息页面对应的结构化信息,在结构化信息卡片的预设位置处添加该结构化信息,生成该信息页面对应的目标搜索结果;所述信息页面的属性标签为该信息页面对应的目标搜索结果的属性标签。
第五方面,本公开实施例还提供一种计算机设备,包括:处理器、存储器和总线,所述存储器存储有所述处理器可执行的机器可读指令,当计算机设备运行时,所述处理器与所述存储器之间通过总线通信,所述机器可读指令被所述处理器执行时执行上述第一方面,或第一方面中任一种可能的实施方式中的步骤,或执行上述第二方面,或第二方面中任一种可能的实施方式中的步骤。
第六方面,本公开实施例还提供一种计算机可读存储介质,该计算机可读存储介质上存储有计算机程序,该计算机程序被处理器运行时执行上述第一方面,或第一方面中任一种可能的实施方式中的步骤,或执行上述第二方面,或第二方面中任一种可能的实施方式中的步骤。
本公开实施例提供的信息展示方法、信息搜索方法及装置,可以在获取到搜索信息对应的多个目标搜索结果之后,根据目标搜索结果对应的属性标签,将各个目标搜索结果展示在对应的属性标签下,从而实现对于目标搜索结果的分类展示,通过这种方法,可以直观的向用户展示不同的属性标签,用户可以选中自己感兴趣的属性标签,从而查看该属性标签下的目标搜索结果,一方面满足了用户对不同属性标签下的目标搜索结果的查看需求,另一方面便于用户快速定位到自己感兴趣的目标搜索结果,从而提高了搜索效率。
为使本公开的上述目的、特征和优点能更明显易懂,下文特举较佳实施例,并配合所附附图,作详细说明如下。
附图说明
为了更清楚地说明本公开实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,此处的附图被并入说明书中并构成本说明书中的一部分,这些附图示出 了符合本公开的实施例,并与说明书一起用于说明本公开的技术方案。应当理解,以下附图仅示出了本公开的某些实施例,因此不应被看作是对范围的限定,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他相关的附图。
图1示出了本公开实施例所提供的一种信息展示方法的流程图;
图2a示出了本公开实施例所提供的一种搜索结果展示页面的示意图;
图2b示出了本公开实施例所提供的另一种搜索结果展示页面的示意图;
图3示出了本公开实施例所提供的另一种搜索结果展示页面的示意图;
图4示出了本公开实施例所提供的一种信息搜索方法的流程示意图;
图5示出了本公开实施例所提供的一种目标搜索结果,以及每一个所述目标搜索结果对应的属性标签的确定方法的示意图;
图6示出了本公开实施例所提供的一种信息展示装置的架构示意图;
图7示出了本公开实施例所提供的一种信息搜索装置的架构示意图;
图8示出了本公开实施例所提供的计算机设备800的结构示意图;
图9示出了本公开实施例所提供的计算机设备900的结构示意图。
具体实施方式
为使本公开实施例的目的、技术方案和优点更加清楚,下面将结合本公开实施例中附图,对本公开实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本公开一部分实施例,而不是全部的实施例。通常在此处附图中描述和示出的本公开实施例的组件可以以各种不同的配置来布置和设计。因此,以下对在附图中提供的本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围,而是仅仅表示本公开的选定实施例。基于本公开的实施例,本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例,都属于本公开保护的范围。
相关技术中,在进行没事内容的搜索结果的展示时,是通过依次展示的方法进行展示,然而这种展示方法可能无法满足用户的搜索需求,用户可能还需要基于其他的搜索词再次发起搜索,例如用户的搜索请求为“宫保鸡丁”,而用户真正感兴趣的为“宫保鸡丁的家常做法”,而其感兴趣的搜索结果可能排在后面,用户无法在当前页面查看到“宫保鸡丁的家常做法”对应的搜索结果,则用户可能会重新以“宫保鸡丁的家常做法”再次发起请求,这就导致搜索效率较低。
基于此,本公开提供了一种信息展示方法、信息搜索方法及装置,可以在获取到搜索信息对应的多个目标搜索结果之后,根据目标搜索结果对应的属性标签,将各个目标搜索结果展示在对应的属性标签下,从而实现对于目标搜索结果的分类展示,通过这种 方法,可以直观的向用户展示不同的属性标签,用户可以选中自己感兴趣的属性标签,从而查看该属性标签下的目标搜索结果,一方面满足了用户对不同属性标签下的目标搜索结果的查看需求,另一方面便于用户快速定位到自己感兴趣的目标搜索结果,从而提高了搜索效率。
针对以上方案所存在的缺陷,均是发明人在经过实践并仔细研究后得出的结果,因此,上述问题的发现过程以及下文中本公开针对上述问题所提出的解决方案,都应该是发明人在本公开过程中对本公开做出的贡献。
应注意到:相似的标号和字母在下面的附图中表示类似项,因此,一旦某一项在一个附图中被定义,则在随后的附图中不需要对其进行进一步定义和解释。
为便于对本实施例进行理解,首先对本公开实施例所公开的一种信息展示方法进行详细介绍,本公开实施例所提供的信息展示方法的执行主体一般为具有终端设备,终端设备可以为智能手机、平板电脑、智能电视、个人计算机等。
参见图1所示,为本公开实施例提供的一种信息展示方法的流程图,所述方法包括步骤101~步骤103,其中:
步骤101、接收用户输入的搜索信息。
步骤102、获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息。
步骤103、在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
以下是对上述步骤101~步骤103的详细描述。
针对步骤101、
用户终端在接收用户输入的搜索信息之后,可以基于搜索信息生成搜索请求,并将生成的搜索请求发送至对应的服务器中,其中,所述搜索请求中可以携带所述搜索信息。
针对步骤102、
在获取搜索信息对应的多个目标搜索结果和多个属性标签之后,确定每个属性标签下的至少一个目标搜索结果时,可以通过以下方法中的任意一种方法:
方法一、
获取的多个属性标签与多个目标搜索结果之间具有映射关系,在确定每个属性标签下的至少一个目标搜索结果时,可以直接根据多个属性标签与多个目标搜索结果之间的映射关系,确定每个属性标签下的至少一个目标搜索结果。
这里,所述映射关系的具体表现形式可以为属性标签与各个目标搜索结果的标识信息之间的映射关系,其映射关系可以是由服务器在确定各个目标搜索结果之后,由服务器生成的。
示例性的,其映射关系可以如下表1所示:
表1
属性标签 目标搜索结果
1 a、b、c、d
2 e、f、g、h
方法二、
每个目标搜索结果都携带有属性标签,在确定每个属性标签下的至少一个属性标签,在确定每个属性标签下的至少一个目标搜索结果时,可以基于目标搜索结果携带的属性标签,对所述多个目标搜索结果进行聚类,确定每个所述属性标签下的至少一个目标搜索结果。
具体实施中,在进行聚类时,可以将携带相同属性标签的目标搜索结果划分至同一类,在将全部目标搜索结果划分完之后,可以确定每个属性标签下的至少一个目标搜索结果。
方法三、
每个目标搜索携带有至少一个属性标签,以及该所述目标搜索结果在各所述属性标签下的质量分数,所述质量分数用于表示所述目标搜索结果与该属性标签之间的匹配程度,其中,所述目标搜索结果对应的质量分数是由服务器确定的。
在确定每个所述属性标签下的至少一个目标搜索结果时,针对每个目标搜索结果,可以基于该目标搜索结果在携带的各所述属性标签下的质量分数,确定该目标搜索结果对应的目标属性标签;然后基于每个目标搜索结果对应的目标属性标签,对所述多个目标搜索结果进行聚类,确定每个属性标签下的至少一个目标搜索结果。
在一种可能的实施方式中,在确定每个目标搜索结果对应的目标属性标签时,可以将该目标搜索结果携带的各个属性标签中,对应的质量分数最高的属性标签作为该目标搜索结果对应的目标属性标签。
针对步骤103、
在一种可能的实施方式中,在搜索结果展示页面展示多个属性标签、以及属性标签下的至少一个目标搜索结果时,可以在搜索结果展示页面横向展示多个属性标签,然后从属性标签中选择第一属性标签,在所述第一属性标签下横向展示所述第一属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示,示例性的,搜索结果展示页面可以如图2a所示。
其中,在横向展示多个属性标签时,可以是基于预先设置好的属性标签展示顺序进行展示,在一种可能的实施方式中,搜索信息为菜谱搜索信息,属性标签用于指示菜谱属性信息,示例性的属性标签可以包括家常、下饭菜、简单易做、名家等。
在确定第一属性标签时,可以是在接收到针对属性标签的第一触发操作之后,将第 一触发操作对应的属性标签作为第一属性标签,或者,第一属性标签可以是展示顺序排在首位的属性标签。
每个属性标签下可以对应至少一个目标搜索结果,在搜索结果展示页面中展示第一属性标签下的至少一个目标搜索结果时,可以根据至少一个目标搜索结果对应的质量分数进行排序,并排序结果并排展示所述至少一个目标搜索结果。
在另一种可能的实施方式中,在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果时,可以在所述搜索结果展示页面横向展示所述多个属性标签;针对各所述属性标签,在该属性标签下纵向展示该属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示,示例性的,搜索结果展示页面可以如图2b所示。
这里需要说明的是,在搜索结果展示页面中除了展示目标搜索结果外,还展示除目标搜索结果外的其他自然搜索结果,这里,至少一个目标搜索结果中的结构化信息卡片支持滑动展示,通过上下滑动,可以实现同一属性标签下的多个结构化信息卡片的切换展示,当用户想要查看除目标搜索结果外的其他自然搜索结果时,可以通过用户的滑动距离进行确定。
示例性的,若用户在用户端的滑动距离小于第一预设距离时,可以切换在当前属性标签下展示的结构化信息卡片,若用户在用户端的滑动距离大于或等于第一预设距离时,可以直接将搜索结果展示页面进行向上滑动,展示其他自然搜索结果。
在一种可能的应用场景中,目标搜索结果中的结构化信息可以包含目标菜谱的菜谱多媒体信息和菜谱描述信息,所述菜谱描述信息中包含食材信息和制作流程;所述菜谱评价信息可以包括评分、人气值、收藏数、以及制作耗时等,所述制作流程可以包括制作视频和制作步骤,在结构化信息卡片中以预设形式展示所述结构化信息。
所述菜谱多媒体信息可以是以多媒体的形式展示所述菜谱的信息,例如可以包括菜谱图片、和/或菜谱视频等,所述菜谱视频可以是用视频形式来展示所述菜谱的视频。
示例性的,如图2b所示,在结构化信息生成对应的结构化信息卡片时,可以在结构化信息卡片的第一预设位置处展示菜谱多媒体信息,然后在结构化信息卡片的第二预设位置处展示菜谱描述信息。
示例性的,搜索结果展示页面如图3所示,图3中包括“家常、下饭菜、简单易做、名家”四个属性标签,第一属性标签为“家常”,图3页面中当前展示的为第一属性标签下质量分数最高的搜索结果,该搜索结果中包括制作视频、食材、制作步骤(即图中的“做法”)。
在接收各个目标搜索结果之后,初始展示时,可以展示上述多个属性标签,并在第一属性标签下展示该属性标签下质量分数最高的目标搜索结果,在一种可能的实施方式中,还可以响应于针对除第一属性标签外的其他属性标签的触发操作,展示该触发操作对应的第二属性标签下质量分数最高的目标搜索结果。
其中,所述针对除第一属性标签外的其他属性标签的触发操作可以包括单击、双击、长按、重按等。
示例性的,若当前展示页面如图3所示,用户点击“下饭菜”的属性标签之后,可以展示在“下饭菜”的属性标签下质量分数最高的目标搜索结果。
在另外一种可能的实施方式中,还可以响应于针对除质量分数最高的目标搜索结果外的其他目标搜索结果的滑动操作,展示其他目标搜索结果。
示例性的,所述针对除质量分数最高的目标搜索结果外的其他目标搜索结果的滑动操作,若当前展示页面如图3所示,用户可以通过左右滑动制作视频所在区域,从而实现对于“家常”属性标签下其他目标搜索结果的展示。
在一种可能的实施方式中,还可以响应于针对任一目标搜索结果的第二触发操作,由搜索结果展示页面跳转至所述触发操作对应的目标搜索结果的信息页面,其中,该目标搜索结果中的结构化信息从该信息页面中提取。
目标搜索结果的信息页面中可以展示有该目标搜索结果对应的视频、图片、文字、评论等信息。
另外,对于其他可以进行步骤详解的搜索信息本公开所提供的方法也可以适用,例如所述搜索信息可以是折纸教程搜索信息,此时,折纸教程搜索结果对应的属性标签可以包括简单易学、精美等,该属性标签为预先设置好的。
本公开实施例提供的信息展示方法,可以在获取到搜索信息对应的多个目标搜索结果之后,根据目标搜索结果对应的属性标签,将各个目标搜索结果展示在对应的属性标签下,从而实现对于目标搜索结果的分类展示,通过这种方法,可以直观的向用户展示不同属性标签下的目标搜索内容,用户可以基于展示的属性标签的选择实现对于不同目标搜索结果的查看,满足了用户基于不同属性标签下的搜索需求,提高了搜索效率。
参见图4所示,为本公开实施例提供的一种信息搜索方法的流程示意图,该方法可以应用于服务器,包括以下几个步骤:
步骤401、获取搜索信息。
步骤402、基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系。
步骤403、根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息。
步骤404、将所述多个目标搜索结果和所述多个属性标签发送至用户端。
以下是对上述步骤401~步骤404的详细介绍。
针对步骤401、
其中,所述获取搜索信息可以是获取用户端发送的搜索请求,该搜索请求中携带有用户输入的搜索信息。
针对步骤402、
具体的,所述索引库中存储有多个信息页面、从每个信息页面中提取的结构化信息、以及该信息页面对应的属性标签,所述信息页面与结构化信息、属性标签之间的映射关系,可以是信息页面标识、结构化信息标识、以及属性标签之间的映射关系。
在一种可能的实施方式中,在基于索引库确定与所述搜索信息匹配的多个信息页面时,可以通过如图5所示的方法,包括以下几个步骤:
步骤501、对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面;其中,每个待筛选信息页面对应有属性标签。
其中,在基于判定结果,确定与所述搜索信息对应的多个待筛选信息页面时,可以先对所述搜索信息进行分词处理,确定所述搜索信息中包含的至少一个搜索关键词;然后获取与所述搜索关键词相对应的至少一个同义词;再基于所述搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结果对应的索引库中,查找与所述搜索信息对应的多个待筛选信息页面,其中,所述索引库可以是倒排索引库。
倒排索引库中存储有多条索引信息;不同的识别结果对应不同的倒排索引库,例如若搜索信息的识别结果为菜谱信息搜索,则可以从菜谱倒排索引库中查找与搜索信息对应的多个待筛选信息页面,若搜索信息的识别结果为折纸教程搜索,则可以从折纸教程倒排索引库中查找与搜索信息对应的多个待筛选信息页面。
在一种可能的实施方式中,倒排索引库中每一条索引信息都对应有唯一的索引关键词,基于该索引关键词可以直接检索到该索引信息,在基于搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结果对应的倒排索引库中,查找与所述搜索信息对应的多个待筛选信息页面时,可以分别计算所述搜索关键词以及所述搜索关键词相对的同义词与各个索引信息对应的索引关键词之间的相关度,当任一索引信息对应的相关度大于预设相关度时,可以将该索引信息对应的信息页面确定为与搜索信息匹配的待筛选信息页面。
其中,所述倒排索引库中存储的索引信息可以包括索引关键词,以及指示包括所述索引关键词的倒排列表,示例性的其存储形式可以如下表1所示:
表1
索引关键词 倒排列表
宫保鸡丁 (5,1)
鱼香肉丝 (2,1),(5,1)
京酱肉丝 (1,1),(3,3),(4,2)
上述倒排列表中(x,y)表示第x个信息页面的第y个位置处包括索引关键词,例如1可以表示标题部分,2表示摘要部分,3表示正文部分,对于索引关键词“京酱肉丝”,在第1个信息页面的标题部分、第3个信息页面的正文部分、以及第4个信息页面的摘要部分均包含该索引关键词。
基于搜索关键词、所述搜索关键词相对应的至少一个同义词,可以查找到对应的倒排列表,然后对查找到的倒排列表进行归并处理,归并处理之后的列表中的信息页面作为搜索信息对应的待筛选信息页面。
其中,倒排索引库中还存储有多个信息页面对应的结构化信息,每个信息页面都有对应的预先设置的属性标签,所述信息页面对应的结构化信息可以是该从该信息页面中提取的结构化信息。
示例性的,对于索引关键词“宫保鸡丁”,可以有与该索引关键词对应的多个信息页面,然后可以选取对应的点击次数较多的信息页面,并从选取出的信息页面中提取结构化信息,作为索引关键词“宫保鸡丁”对应的信息页面,以及该信息页面对应的结构化信息。
其中,在从信息页面中提取结构化信息时,可以分别从标题、摘要、简介、食材、步骤、作者信息等多个内容进行提取,示例性的,提取的结构化信息可以如下表2所示:
表2
Figure PCTCN2021093571-appb-000001
上表2中,详细信息一栏可以所填写的内容可以是从信息页面中提取得到的,这里,由于所需提取的内容是固定的,在进行搜索结果的展示时,可以将各个结构化信息按照预设形式展示在结构化信息卡片中。
在一种可能的实施方式中,索引库中在存储时,可以预先存储每个信息页面对应的属性标签,具体的,可以预先根据以下方法确定每个待筛选信息页面对应的属性标签:
预先设置每个属性标签对应的至少一个标签关键词;
从所述待筛选信息页面中查找所述标签关键词,若查找到任一标签关键词,将查找到的标签关键词对应的属性标签作为该待筛选信息页面的属性标签。
示例性的,对于属性标签“家常”可以设置对应的标签关键词为家庭做法、家庭详细步骤等,对于属性标签“名家”,可以设置对应的标签关键词为作者的名称。
不同属性标签在索引文件中的查找位置不同,例如对于属性标签“家常”,可以从索引文件的标题部分和简介部分查找,对于属性标签“下饭菜”,可以从索引文件的用户评论部分查找,对于属性标签“名家”,可以从索引文件的作者名称部分查找。
其中,若索引文件中并未查找到任何标签关键词,则可以基于人工去为该索引文件 添加属性标签。
在另外一种可能的实施方式中,在确定每个待筛选信息页面对应的属性标签时,还可以将待筛选信息页面输入至预先训练的神经网络中,神经网络可以输出该待筛选信息页面对应的属性标签,其中,所述神经网络是基于样本文档训练得到的,所述样本文档有对应的预先标注的属性标签。
这里需要说明的是,倒排索引库中所存储的索引文件可以仅有一个属性标签,也可以有多个属性标签。
具体实施中,对所述搜索信息进行搜索类型判定时,可以基于训练的判定模型和所述搜索信息,确定所述搜索信息对应的判定结果,其中,所述判定模型是基于样本信息,以及所述样本信息携带的搜索类型标签训练得到的。
步骤502、基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面。
在一种可能的实施方式中,在确定所述多个待筛选信息页面的质量分数时,可以先确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度;以及,获取所述待筛选信息页面的文本质量参数;然后将所述文本相关度和所述文本质量参数进行加权求和,得到所述待筛选信息页面的质量分数。其中,所述文本质量参数可以包括图文质量参数和/或图片清晰度。
在确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,可以将所述待筛选信息页面的标题、摘要、正文内容分别对应的词向量输入至训练的语言处理模型,得到所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量;然后基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
其中,所述语言处理模型可以是BERT模型,所述文本相关度可以是用于表示待筛选信息页面的标题、摘要和正文部分之间的关联程度,在一种可能的场景中,待筛选信息页面的标题可能与摘要和正文部分之间的关联程度较低,在这种情况下,若在待筛选信息页面的标题中查找到与任一属性标签对应的标签关键词,但待筛选信息页面的标题与摘要和正文部分之间的关联程度较低,基于这类待筛选信息页面生成的目标搜索结果可能会导致精度较低,因此可以通过文本相关度对待筛选信息页面进行筛选。
一种可选的实施方式中,在基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,可以将待筛选信息页面的标题、摘要、正文内容分别对应的语义向量进行相乘,并将乘积作为待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
在另外一种可能的实施方式中,在基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,将所述待筛选信息页面的摘要、正文内容分别对应的语义向量进行加权求和, 得到第一语义向量;然后将所述待筛选信息页面的标题对应的第二语义向量与所述第一语义向量相乘,得到文本相关度。
在一种可能的实施方式中,在基于所述待筛选信息页面的质量分数和所述待筛选信息页面对应的属性标签,从所述待筛选信息页面中筛选出目标搜索结果时,针对每一个属性标签,可以先确定对应该属性标签的待筛选信息页面,并将对应该属性标签的待筛选信息页面按照对应的质量分数进行排序;然后基于排序结果排在前N位的待筛选信息页面的结构化信息生成目标搜索结果,N为预设正整数。
在一种可能的实施方式中,在基于所述待筛选信息页面的质量分数和所述待筛选信息页面对应的属性标签,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面时,可以先基于所述待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出中间信息页面;然后基于所述中间信息页面的质量分数和所述中间搜索结果对应的属性标签,从所述中间搜索结果中筛选出目标搜索结果。
具体的,在从中间信息页面中筛选出与所述搜索信息匹配的多个信息页面时,可以根据中间信息页面对应的属性标签,确定每一个属性标签对应的中间信息页面,针对每一个属性标签,将该属性标签下的中间信息页面按照对应的质量分数进行排序,然后将排在前M位的中间信息页面作为与所述搜索信息匹配的信息页面,其中,M为预设值。
在一种可能的实施方式中,在基于排序结果,确定与所述搜索信息匹配的信息页面时,不同属性标签下所筛选出的与所述搜索信息匹配的信息页面的个数可以相同,例如可以为三个,在另一种可能的实施方式中,不同属性标签的重要程度可能不同,可以预先设置不同属性标签下对应的与所述搜索信息匹配的信息页面的个数,针对每一个属性标签,在筛选与所述搜索信息匹配的信息页面时,可以按照该属性标签对应的个数进行筛选。
在一种可能的实施方式中,若某一待筛选信息页面对应有多个属性标签,且在每个属性标签下该待筛选信息页面的质量分数都比较高,均可作为每个属性标签下的与所述搜索信息匹配的信息页面,在这种情况下,可以将该待筛选信息页面质量分数最高的属性标签作为该待筛选信息页面所对应的目标属性标签,然后将该待筛选信息页面作为该目标属性标签下的与所述搜索信息匹配的信息页面。
针对步骤403、
在一种可能的实施方式中,目标搜索结果可以包括展示有结构化信息的结构化信息卡片。
在根据多个信息页面对应的结构化信息和属性标签,生成搜索信息对应的多个目标搜索结果和多个属性标签时,可以基于与搜索信息匹配的每一个信息页面对应的结构化信息,在结构化信息卡片的预设位置处添加该结构化信息,生成该信息页面对应的目标搜索结果,其中,所述信息页面的属性标签为该信息页面对应的目标搜索结果的属性标签。
针对步骤404、
在将目标搜索结果,以及目标搜索结果对应的属性标签发送至用户端之后,用户端可以基于如图1所述的信息展示方法,对接收到的目标搜索结果进行展示。
另外,服务器在向用户端发送目标搜索结果时,还可以将其他自然搜索结果发送至用户端,所述自然搜索结果为服务器基于该搜索信息查找到的普通搜索结果,目标搜索结果与之不同的是,目标搜索结果是基于目标搜索结果对应的倒排索引库中查找到的,而所述普通搜索结果是服务器基于所有可获取搜索结果的数据库中查找到的。
本公开实施例提供的信息展示方法、信息搜索方法及装置,可以在获取到搜索信息对应的多个目标搜索结果之后,根据目标搜索结果对应的属性标签,将各个目标搜索结果展示在对应的属性标签下,从而实现对于目标搜索结果的分类展示,通过这种方法,可以直观的向用户展示不同的属性标签,用户可以选中自己感兴趣的属性标签,从而查看该属性标签下的目标搜索结果,一方面满足了用户对不同属性标签下的目标搜索结果的查看需求,另一方面便于用户快速定位到自己感兴趣的目标搜索结果,从而提高了搜索效率。
本领域技术人员可以理解,在具体实施方式的上述方法中,各步骤的撰写顺序并不意味着严格的执行顺序而对实施过程构成任何限定,各步骤的具体执行顺序应当以其功能和可能的内在逻辑确定。
基于同一发明构思,本公开实施例中还提供了与信息展示方法对应的信息展示装置,由于本公开实施例中的装置解决问题的原理与本公开实施例上述信息展示方法相似,因此装置的实施可以参见方法的实施,重复之处不再赘述。
参照图6所示,为本公开实施例提供的一种信息展示装置的架构示意图,所述装置包括:接收模块601、第一确定模块602、展示模块603;其中,
接收模块601,用于接收用户输入的搜索信息;
第一确定模块602,用于获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息;
展示模块603,用于在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,获取的所述多个属性标签与所述多个目标搜索结果具有映射关系;
所述第一确定模块602,在确定每个所述属性标签下的至少一个目标搜索结果时,用于:
基于所述映射关系确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,每个所述目标搜索结果携带有至少一个属性标签;
所述第一确定模块602,在确定每个所述属性标签下的至少一个目标搜索结果时, 用于:
基于所述目标搜索结果携带的至少一个属性标签,对所述多个目标搜索结果进行聚类,确定每个所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,每个所述目标搜索结果携带有至少一个属性标签,以及该所述目标搜索结果在各所述属性标签下的质量分数,所述质量分数用于表示所述目标搜索结果与该属性标签之间的匹配程度。
一种可能的实施方式中,所述第一确定模块602,在确定每个所述属性标签下的至少一个目标搜索结果时,用于:
针对每个目标搜索结果,基于该目标搜索结果在携带的各所述属性标签下的质量分数,确定该目标搜索结果对应的目标属性标签;
基于每个目标搜索结果对应的目标属性标签,对所述多个目标搜索结果进行聚类,确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,所述展示模块603,在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果时,用于:
在所述搜索结果展示页面横向展示所述多个属性标签;
从所述属性标签中选择第一属性标签,在所述第一属性标签下横向展示所述第一属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,所述第一确定模块602,还用于:
接收针对于所述属性标签的第一触发操作,基于所述第一触发操作确定所述第一属性标签。
一种可能的实施方式中,所述展示模块603,在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果时,用于:
在所述搜索结果展示页面横向展示所述多个属性标签;
针对各所述属性标签,在该属性标签下纵向展示该属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,所述搜索信息为菜谱搜索信息,所述属性标签用于指示菜谱的属性信息。
一种可能的实施方式中,所述目标搜索结果中的结构化信息包含目标菜谱的菜谱多媒体信息和菜谱描述信息,所述菜谱描述信息中包含食材信息和制作流程,在所述结构化信息卡片中以预设形式展示所述结构化信息。
一种可能的实施方式中,所述展示模块603,还用于:
响应于对于任一目标搜索结果的第二触发操作,由所述搜索结果展示页面跳转至所述触发操作对应的目标搜索结果的信息页面,其中,该目标搜索结果中的结构化信息从 该信息页面中提取。
基于同一发明构思,本公开实施例中还提供了与信息搜索方法对应的信息搜索装置,由于本公开实施例中的装置解决问题的原理与本公开实施例上述信息搜索方法相似,因此装置的实施可以参见方法的实施,重复之处不再赘述。
参照图7所示,为本公开实施例提供的一种信息搜索装置的架构示意图,所述装置包括:获取模块701、第二确定模块702、生成模块703、以及发送模块704,具体的:
获取模块701,用于获取搜索信息;
第二确定模块702,用于基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系;
生成模块703,用于根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息;
发送模块704,用于将所述多个目标搜索结果和所述多个属性标签发送至用户端。
一种可能的实施方式中,所述第二确定模块702,在基于索引库确定与所述搜索信息匹配的多个信息页面时,用于:
对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面;其中,每个待筛选信息页面对应有属性标签;
基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,所述第二确定模块702,在对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面时,用于:
对所述搜索信息进行分词处理,确定所述搜索信息中包含的至少一个搜索关键词;
获取与所述搜索关键词相对应的至少一个同义词;
基于所述搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结果对应的索引库中,查找与所述搜索信息对应的多个待筛选信息页面。
一种可能的实施方式中,所述第二确定模块702,还用于根据以下方法确定每个待筛选信息页面对应的属性标签:
预先设置每个属性标签对应的至少一个标签关键词;
从所述待筛选信息页面中查找所述标签关键词,若查找到任一标签关键词,将查找到的标签关键词对应的属性标签作为该待筛选信息页面的属性标签。
一种可能的实施方式中,所述第二确定模块702,在对所述搜索信息进行搜索类型判定时,用于:
基于训练的判定模型和所述搜索信息,确定所述搜索信息对应的判定结果,其中, 所述判定模型是基于样本信息,以及所述样本信息携带的搜索类型标签训练得到的。
一种可能的实施方式中,所述第二确定模块702,用于通过以下方法确定所述多个待筛选信息页面的质量分数:
确定所述待筛选信息页面中的标题、摘要、正文内容之间的文本相关度;以及,获取所述待筛选信息页面的文本质量参数;
将所述文本相关度和所述文本质量参数进行加权求和,得到所述待筛选信息页面的质量分数。
一种可能的实施方式中,所述第二确定模块702,在确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,用于:
将所述待筛选信息页面的标题、摘要、正文内容分别对应的词向量输入至训练的语言处理模型,得到所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量;
基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
一种可能的实施方式中,所述第二确定模块702,在基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度时,用于:
将所述待筛选信息页面的摘要、正文内容分别对应的语义向量进行加权求和,得到第一语义向量;
将所述待筛选信息页面的标题对应的第二语义向量与所述第一语义向量相乘,得到文本相关度。
一种可能的实施方式中,所述文本质量参数,包括图文质量参数和/或图片清晰度。
一种可能的实施方式中,所述第二确定模块702,在基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面时,用于:
基于所述待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出中间信息页面;
基于所述中间信息页面的质量分数和所述中间信息页面对应的属性标签,从所述中间信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,所述第二确定模块702,在基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面时,用于:
针对每一个属性标签,确定对应该属性标签的待筛选信息页面,并将对应该属性标签的待筛选信息页面按照对应的质量分数进行排序;
将排序结果排在前N位的待筛选信息页面作为与所述搜索信息匹配的多个信息页面,N为预设正整数。
一种可能的实施方式中,所述目标搜索结果包括展示有所述结构化信息的结构化信息卡片;
所述生成模块703,在根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签时,用于:
基于与所述搜索信息匹配的每一个信息页面对应的结构化信息,在结构化信息卡片的预设位置处添加该结构化信息,生成该信息页面对应的目标搜索结果;所述信息页面的属性标签为该信息页面对应的目标搜索结果的属性标签。
关于装置中的各模块的处理流程、以及各模块之间的交互流程的描述可以参照上述方法实施例中的相关说明,这里不再详述。
基于同一技术构思,本公开实施例还提供了一种计算机设备。参照图8所示,为本公开实施例提供的计算机设备800的结构示意图,包括处理器801、存储器802、和总线803。其中,存储器802用于存储执行指令,包括内存8021和外部存储器8022;这里的内存8021也称内存储器,用于暂时存放处理器801中的运算数据,以及与硬盘等外部存储器8022交换的数据,处理器801通过内存8021与外部存储器8022进行数据交换,当计算机设备800运行时,处理器801与存储器802之间通过总线803通信,使得处理器801在执行以下指令:
接收用户输入的搜索信息;
获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息;
在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,处理器801执行的指令中,获取的所述多个属性标签与所述多个目标搜索结果具有映射关系;
所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
基于所述映射关系确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,处理器801执行的指令中,每个所述目标搜索结果携带有至少一个属性标签;
所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
基于所述目标搜索结果携带的至少一个属性标签,对所述多个目标搜索结果进行聚类,确定每个所述属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,处理器801执行的指令中,每个所述目标搜索结果携带有至少一个属性标签,以及该所述目标搜索结果在各所述属性标签下的质量分数,所述质量分数用于表示所述目标搜索结果与该属性标签之间的匹配程度。
一种可能的实施方式中,处理器801执行的指令中,所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
针对每个目标搜索结果,基于该目标搜索结果在携带的各所述属性标签下的质量分数,确定该目标搜索结果对应的目标属性标签;
基于每个目标搜索结果对应的目标属性标签,对所述多个目标搜索结果进行聚类,确定每个属性标签下的至少一个目标搜索结果。
一种可能的实施方式中,处理器801执行的指令中,所述在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果,包括:
在所述搜索结果展示页面横向展示所述多个属性标签;
从所述属性标签中选择第一属性标签,在所述第一属性标签下横向展示所述第一属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,处理器801执行的指令中,所述方法还包括:
接收针对于所述属性标签的第一触发操作,基于所述第一触发操作确定所述第一属性标签。
一种可能的实施方式中,处理器801执行的指令中,所述在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果,包括:
在所述搜索结果展示页面横向展示所述多个属性标签;
针对各所述属性标签,在该属性标签下纵向展示该属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
一种可能的实施方式中,处理器801执行的指令中,所述搜索信息为菜谱搜索信息,所述属性标签用于指示菜谱的属性信息。
一种可能的实施方式中,处理器801执行的指令中,所述目标搜索结果中的结构化信息包含目标菜谱的菜谱多媒体信息和菜谱描述信息,所述菜谱描述信息中包含食材信息和制作流程,在所述结构化信息卡片中以预设形式展示所述结构化信息。
一种可能的实施方式中,处理器801执行的指令中,所述方法还包括:
响应于对于任一目标搜索结果的第二触发操作,由所述搜索结果展示页面跳转至所述触发操作对应的目标搜索结果的信息页面,其中,该目标搜索结果中的结构化信息从该信息页面中提取。
基于同一技术构思,本公开实施例还提供了一种计算机设备。参照图9所示,为本公开实施例提供的计算机设备900的结构示意图,包括处理器901、存储器902、和总线903。其中,存储器902用于存储执行指令,包括内存9021和外部存储器9022;这里的内存9021也称内存储器,用于暂时存放处理器901中的运算数据,以及与硬盘等外部存储器9022交换的数据,处理器901通过内存9021与外部存储器9022进行数据交换,当计算机设备900运行时,处理器901与存储器902之间通过总线903通信,使得处理器901在执行以下指令:
获取搜索信息;
基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系;
根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息;
将所述多个目标搜索结果和所述多个属性标签发送至用户端。
一种可能的实施方式中,处理器901执行的指令中,所述基于索引库确定与所述搜索信息匹配的多个信息页面,包括:
对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面;其中,每个待筛选信息页面对应有属性标签;
基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,处理器901执行的指令中,所述对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面,包括:
对所述搜索信息进行分词处理,确定所述搜索信息中包含的至少一个搜索关键词;
获取与所述搜索关键词相对应的至少一个同义词;
基于所述搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结果对应的索引库中,查找与所述搜索信息对应的多个待筛选信息页面。
一种可能的实施方式中,处理器901执行的指令中,根据以下方法确定每个待筛选信息页面对应的属性标签:
预先设置每个属性标签对应的至少一个标签关键词;
从所述待筛选信息页面中查找所述标签关键词,若查找到任一标签关键词,将查找到的标签关键词对应的属性标签作为该待筛选信息页面的属性标签。
一种可能的实施方式中,处理器901执行的指令中,所述对所述搜索信息进行搜索类型判定,包括:
基于训练的判定模型和所述搜索信息,确定所述搜索信息对应的判定结果,其中,所述判定模型是基于样本信息,以及所述样本信息携带的搜索类型标签训练得到的。
一种可能的实施方式中,处理器901执行的指令中,通过以下方法确定所述多个待筛选信息页面的质量分数:
确定所述待筛选信息页面中的标题、摘要、正文内容之间的文本相关度;以及,获取所述待筛选信息页面的文本质量参数;
将所述文本相关度和所述文本质量参数进行加权求和,得到所述待筛选信息页面的质量分数。
一种可能的实施方式中,处理器901执行的指令中,所述确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度,包括:
将所述待筛选信息页面的标题、摘要、正文内容分别对应的词向量输入至训练的语言处理模型,得到所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量;
基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
一种可能的实施方式中,处理器901执行的指令中,所述基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度,包括:
将所述待筛选信息页面的摘要、正文内容分别对应的语义向量进行加权求和,得到第一语义向量;
将所述待筛选信息页面的标题对应的第二语义向量与所述第一语义向量相乘,得到文本相关度。
一种可能的实施方式中,处理器901执行的指令中,所述文本质量参数,包括图文质量参数和/或图片清晰度。
一种可能的实施方式中,处理器901执行的指令中,所述基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面,包括:
基于所述待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出中间信息页面;
基于所述中间信息页面的质量分数和所述中间信息页面对应的属性标签,从所述中间信息页面中筛选出与所述搜索信息匹配的多个信息页面。
一种可能的实施方式中,处理器901执行的指令中,所述基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面,包括:
针对每一个属性标签,确定对应该属性标签的待筛选信息页面,并将对应该属性标签的待筛选信息页面按照对应的质量分数进行排序;
将排序结果排在前N位的待筛选信息页面作为与所述搜索信息匹配的多个信息页面,N为预设正整数。
一种可能的实施方式中,处理器901执行的指令中,所述目标搜索结果包括展示有所述结构化信息的结构化信息卡片;
所述根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签,包括:
基于与所述搜索信息匹配的每一个信息页面对应的结构化信息,在结构化信息卡片的预设位置处添加该结构化信息,生成该信息页面对应的目标搜索结果;所述信息页面 的属性标签为该信息页面对应的目标搜索结果的属性标签。
本公开实施例还提供一种计算机可读存储介质,该计算机可读存储介质上存储有计算机程序,该计算机程序被处理器运行时执行上述方法实施例中所述的信息展示方法、信息搜索方法的步骤。其中,该存储介质可以是易失性或非易失的计算机可读取存储介质。
本公开实施例所提供的信息展示方法、信息搜索方法的计算机程序产品,包括存储了程序代码的计算机可读存储介质,所述程序代码包括的指令可用于执行上述方法实施例中所述的信息展示方法、信息搜索方法的步骤,具体可参见上述方法实施例,在此不再赘述。
本公开实施例还提供一种计算机程序,该计算机程序被处理器执行时实现前述实施例的任意一种方法。该计算机程序产品可以具体通过硬件、软件或其结合的方式实现。在一个可选实施例中,所述计算机程序产品具体体现为计算机存储介质,在另一个可选实施例中,计算机程序产品具体体现为软件产品,例如软件开发包(Software Development Kit,SDK)等等。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统和装置的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。在本公开所提供的几个实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,又例如,多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本公开各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解,本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、 磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是:以上所述实施例,仅为本公开的具体实施方式,用以说明本公开的技术方案,而非对其限制,本公开的保护范围并不局限于此,尽管参照前述实施例对本公开进行了详细的说明,本领域的普通技术人员应当理解:任何熟悉本技术领域的技术人员在本公开揭露的技术范围内,其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化,或者对其中部分技术特征进行等同替换;而这些修改、变化或者替换,并不使相应技术方案的本质脱离本公开实施例技术方案的精神和范围,都应涵盖在

Claims (27)

  1. 一种信息展示方法,其特征在于,包括:
    接收用户输入的搜索信息;
    获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息;
    在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
  2. 根据权利要求1所述的方法,其特征在于,获取的所述多个属性标签与所述多个目标搜索结果具有映射关系;
    所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
    基于所述映射关系确定每个属性标签下的至少一个目标搜索结果。
  3. 根据权利要求1所述的方法,其特征在于,每个所述目标搜索结果携带有属性标签;
    所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
    基于所述目标搜索结果携带的属性标签,对所述多个目标搜索结果进行聚类,确定每个所述属性标签下的至少一个目标搜索结果。
  4. 根据权利要求1所述的方法,其特征在于,每个所述目标搜索结果携带有至少一个属性标签,以及该所述目标搜索结果在各所述属性标签下的质量分数,所述质量分数用于表示所述目标搜索结果与该属性标签之间的匹配程度。
  5. 根据权利要求4所述的方法,其特征在于,所述确定每个所述属性标签下的至少一个目标搜索结果,包括:
    针对每个目标搜索结果,基于该目标搜索结果在携带的各所述属性标签下的质量分数,确定该目标搜索结果对应的目标属性标签;
    基于每个目标搜索结果对应的目标属性标签,对所述多个目标搜索结果进行聚类,确定每个属性标签下的至少一个目标搜索结果。
  6. 根据权利要求1所述的方法,其特征在于,所述在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果,包括:
    在所述搜索结果展示页面横向展示所述多个属性标签;
    从所述属性标签中选择第一属性标签,在所述第一属性标签下横向展示所述第一属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
  7. 根据权利要求6所述的方法,其特征在于,所述方法还包括:
    接收针对于所述属性标签的第一触发操作,基于所述第一触发操作确定所述第一属性标签。
  8. 根据权利要求1所述的方法,其特征在于,所述在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果,包括:
    在所述搜索结果展示页面横向展示所述多个属性标签;
    针对各所述属性标签,在该属性标签下纵向展示该属性标签对应的至少一个目标搜索结果中的结构化信息卡片,所述至少一个目标搜索结果中的结构化信息卡片支持滑动展示。
  9. 根据权利要求6~8任一所述的方法,其特征在于,所述搜索信息为菜谱搜索信息,所述属性标签用于指示菜谱的属性信息。
  10. 根据权利要求9所述的方法,其特征在于,所述目标搜索结果中的结构化信息包含目标菜谱的菜谱多媒体信息和菜谱描述信息,所述菜谱描述信息中包含食材信息和制作流程,在所述结构化信息卡片中以预设形式展示所述结构化信息。
  11. 根据权利要求1所述的方法,其特征在于,所述方法还包括:
    响应于对于任一目标搜索结果的第二触发操作,由所述搜索结果展示页面跳转至所述触发操作对应的目标搜索结果的信息页面,其中,该目标搜索结果中的结构化信息从该信息页面中提取。
  12. 一种信息搜索方法,其特征在于,包括:
    获取搜索信息;
    基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系;
    根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息;
    将所述多个目标搜索结果和所述多个属性标签发送至用户端。
  13. 根据权利要求12所述的方法,其特征在于,所述基于索引库确定与所述搜索信息匹配的多个信息页面,包括:
    对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面;其中,每个待筛选信息页面对应有属性标签;
    基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面。
  14. 根据权利要求13所述的方法,其特征在于,所述对所述搜索信息进行搜索类型判定,并基于判定结果,从与该判定结果对应的索引库中,确定与所述搜索信息匹配的多个待筛选信息页面,包括:
    对所述搜索信息进行分词处理,确定所述搜索信息中包含的至少一个搜索关键词;
    获取与所述搜索关键词相对应的至少一个同义词;
    基于所述搜索关键词、所述搜索关键词相对应的至少一个同义词,从与所述判定结 果对应的索引库中,查找与所述搜索信息对应的多个待筛选信息页面。
  15. 根据权利要求13所述的方法,其特征在于,根据以下方法确定每个待筛选信息页面对应的属性标签:
    预先设置每个属性标签对应的至少一个标签关键词;
    从所述待筛选信息页面中查找所述标签关键词,若查找到任一标签关键词,将查找到的标签关键词对应的属性标签作为该待筛选信息页面的属性标签。
  16. 根据权利要求13所述的方法,其特征在于,所述对所述搜索信息进行搜索类型判定,包括:
    基于训练的判定模型和所述搜索信息,确定所述搜索信息对应的判定结果,其中,所述判定模型是基于样本信息,以及所述样本信息携带的搜索类型标签训练得到的。
  17. 根据权利要求13所述的方法,其特征在于,通过以下方法确定所述多个待筛选信息页面的质量分数:
    确定所述待筛选信息页面中的标题、摘要、正文内容之间的文本相关度;以及,获取所述待筛选信息页面的文本质量参数;
    将所述文本相关度和所述文本质量参数进行加权求和,得到所述待筛选信息页面的质量分数。
  18. 根据权利要求17所述的方法,其特征在于,所述确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度,包括:
    将所述待筛选信息页面的标题、摘要、正文内容分别对应的词向量输入至训练的语言处理模型,得到所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量;
    基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度。
  19. 根据权利要求18所述的方法,其特征在于,所述基于所述待筛选信息页面的标题、摘要、正文内容分别对应的语义向量,确定所述待筛选信息页面的标题、摘要、正文内容之间的文本相关度,包括:
    将所述待筛选信息页面的摘要、正文内容分别对应的语义向量进行加权求和,得到第一语义向量;
    将所述待筛选信息页面的标题对应的第二语义向量与所述第一语义向量相乘,得到文本相关度。
  20. 根据权利要求17所述的方法,其特征在于,所述文本质量参数,包括图文质量参数和/或图片清晰度。
  21. 根据权利要求13所述的方法,其特征在于,所述基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面,包括:
    基于所述待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出中间信息页 面;
    基于所述中间信息页面的质量分数和所述中间信息页面对应的属性标签,从所述中间信息页面中筛选出与所述搜索信息匹配的多个信息页面。
  22. 根据权利要求13所述的方法,其特征在于,所述基于所述多个待筛选信息页面的质量分数,从所述待筛选信息页面中筛选出与所述搜索信息匹配的多个信息页面,包括:
    针对每一个属性标签,确定对应该属性标签的待筛选信息页面,并将对应该属性标签的待筛选信息页面按照对应的质量分数进行排序;
    将排序结果排在前N位的待筛选信息页面作为与所述搜索信息匹配的多个信息页面,N为预设正整数。
  23. 根据权利要求12所述的方法,其特征在于,所述目标搜索结果包括展示有所述结构化信息的结构化信息卡片;
    所述根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签,包括:
    基于与所述搜索信息匹配的每一个信息页面对应的结构化信息,在结构化信息卡片的预设位置处添加该结构化信息,生成该信息页面对应的目标搜索结果;所述信息页面的属性标签为该信息页面对应的目标搜索结果的属性标签。
  24. 一种信息展示装置,其特征在于,包括:
    接收模块,用于接收用户输入的搜索信息;
    第一确定模块,用于获取所述搜索信息对应的多个目标搜索结果和多个属性标签,并确定每个所述属性标签下的至少一个目标搜索结果;其中,所述目标搜索结果包括结构化信息;
    展示模块,用于在搜索结果展示页面展示所述多个属性标签,以及所述属性标签下的至少一个目标搜索结果。
  25. 一种信息搜索装置,其特征在于,包括:
    获取模块,用于获取搜索信息;
    第二确定模块,用于基于索引库确定与所述搜索信息匹配的多个信息页面;其中,所述索引库包括信息页面与结构化信息、属性标签之间的映射关系;
    生成模块,用于根据所述多个信息页面对应的结构化信息和属性标签,生成所述搜索信息对应的多个目标搜索结果和多个属性标签;其中,所述目标搜索结果中包括结构化信息;
    发送模块,用于将所述多个目标搜索结果和所述多个属性标签发送至用户端。
  26. 一种计算机设备,其特征在于,包括:处理器、存储器和总线,所述存储器存储有所述处理器可执行的机器可读指令,当计算机设备运行时,所述处理器与所述存储器之间通过总线通信,所述机器可读指令被所述处理器执行时执行如权利要求1至11 任一所述的信息展示方法,或执行如权利要求12至23任一所述的信息搜索方法的步骤。
  27. 一种计算机可读存储介质,其特征在于,该计算机可读存储介质上存储有计算机程序,该计算机程序被处理器运行时执行如权利要求1至11任一所述的信息展示方法,或执行如权利要求12至23任一所述的信息搜索方法的步骤。
PCT/CN2021/093571 2020-07-31 2021-05-13 一种信息展示方法、信息搜索方法及装置 WO2022022002A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/006,715 US20230252103A1 (en) 2020-07-31 2021-05-13 Method for displaying information, method for searching for information and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010760603.7A CN111859195A (zh) 2020-07-31 2020-07-31 一种信息展示方法、信息搜索方法及装置
CN202010760603.7 2020-07-31

Publications (1)

Publication Number Publication Date
WO2022022002A1 true WO2022022002A1 (zh) 2022-02-03

Family

ID=72953701

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/093571 WO2022022002A1 (zh) 2020-07-31 2021-05-13 一种信息展示方法、信息搜索方法及装置

Country Status (3)

Country Link
US (1) US20230252103A1 (zh)
CN (1) CN111859195A (zh)
WO (1) WO2022022002A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114470751A (zh) * 2022-02-09 2022-05-13 北京有竹居网络技术有限公司 内容获取方法、装置、存储介质及电子设备

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111859195A (zh) * 2020-07-31 2020-10-30 北京字节跳动网络技术有限公司 一种信息展示方法、信息搜索方法及装置
CN112307294A (zh) * 2020-11-02 2021-02-02 北京搜狗科技发展有限公司 一种数据处理方法及装置
CN113821716A (zh) * 2021-01-12 2021-12-21 北京沃东天骏信息技术有限公司 信息搜索方法和装置
CN112948665A (zh) * 2021-01-14 2021-06-11 北京搜狗科技发展有限公司 一种搜索方法、装置和介质
CN113204691B (zh) * 2021-05-31 2023-08-04 抖音视界有限公司 一种信息展示方法、装置、设备及介质
CN113420164A (zh) * 2021-07-19 2021-09-21 北京字节跳动网络技术有限公司 一种信息展示方法、信息搜索方法及装置
CN113486252A (zh) * 2021-07-30 2021-10-08 北京字节跳动网络技术有限公司 搜索结果展示方法、装置、设备和介质
CN113742411A (zh) * 2021-09-06 2021-12-03 科大讯飞股份有限公司 一种信息获取方法、装置、系统和计算机可读存储介质
CN113779307A (zh) * 2021-09-28 2021-12-10 北京字跳网络技术有限公司 一种信息展示方法、装置、计算机设备及存储介质
CN113836434B (zh) * 2021-11-25 2022-03-04 山东捷瑞数字科技股份有限公司 一种基于数据库的web页面数据处理方法
CN114372215B (zh) * 2022-01-12 2023-07-14 抖音视界有限公司 一种搜索结果展示、搜索请求处理方法及装置
CN114756751A (zh) * 2022-04-18 2022-07-15 北京字节跳动网络技术有限公司 推荐词确定方法、装置、电子设备和存储介质
CN115098724A (zh) * 2022-06-24 2022-09-23 抖音视界(北京)有限公司 一种搜索结果展示方法、装置、计算机设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102955807A (zh) * 2011-08-26 2013-03-06 华为软件技术有限公司 一种关联信息的检索方法及装置
CN107656787A (zh) * 2017-09-26 2018-02-02 掌阅科技股份有限公司 基于电子书生成话题的方法、计算设备、计算机存储介质
CN108268582A (zh) * 2017-07-14 2018-07-10 广东神马搜索科技有限公司 信息查询方法及装置
CN111859195A (zh) * 2020-07-31 2020-10-30 北京字节跳动网络技术有限公司 一种信息展示方法、信息搜索方法及装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104035957B (zh) * 2014-04-14 2019-01-25 百度在线网络技术(北京)有限公司 搜索方法和装置
CN105843817A (zh) * 2015-01-15 2016-08-10 北京三星通信技术研究有限公司 在终端设备上进行搜索的方法、装置和设备
CN105893564A (zh) * 2016-03-31 2016-08-24 百度在线网络技术(北京)有限公司 一种基于搜索引擎客户端的搜索方法和装置
CN111209374B (zh) * 2020-01-07 2023-06-06 平安科技(深圳)有限公司 数据查询展示方法、装置、计算机系统及可读存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102955807A (zh) * 2011-08-26 2013-03-06 华为软件技术有限公司 一种关联信息的检索方法及装置
CN108268582A (zh) * 2017-07-14 2018-07-10 广东神马搜索科技有限公司 信息查询方法及装置
CN107656787A (zh) * 2017-09-26 2018-02-02 掌阅科技股份有限公司 基于电子书生成话题的方法、计算设备、计算机存储介质
CN111859195A (zh) * 2020-07-31 2020-10-30 北京字节跳动网络技术有限公司 一种信息展示方法、信息搜索方法及装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114470751A (zh) * 2022-02-09 2022-05-13 北京有竹居网络技术有限公司 内容获取方法、装置、存储介质及电子设备
CN114470751B (zh) * 2022-02-09 2024-02-02 北京有竹居网络技术有限公司 内容获取方法、装置、存储介质及电子设备

Also Published As

Publication number Publication date
CN111859195A (zh) 2020-10-30
US20230252103A1 (en) 2023-08-10

Similar Documents

Publication Publication Date Title
WO2022022002A1 (zh) 一种信息展示方法、信息搜索方法及装置
WO2022048360A1 (zh) 一种搜索结果展示的方法、装置及计算机存储介质
JP7480317B2 (ja) 検索方法、装置、電子機器及び記憶媒体
US20180181569A1 (en) Visual category representation with diverse ranking
US8433698B2 (en) Matching and recommending relevant videos and media to individual search engine results
US8725717B2 (en) System and method for identifying topics for short text communications
US8589429B1 (en) System and method for providing query recommendations based on search activity of a user base
US8380727B2 (en) Information processing device and method, program, and recording medium
US20140172821A1 (en) Generating filters for refining search results
US10909202B2 (en) Information providing text reader
WO2023273686A1 (zh) 一种信息搜索方法、装置、计算机设备及存储介质
WO2022111249A1 (zh) 一种信息展示的方法、装置以及计算机存储介质
EP2368200A2 (en) Interactively ranking image search results using color layout relevance
US11100169B2 (en) Alternative query suggestion in electronic searching
US11755651B2 (en) Method, apparatus, and computer-readable medium for generating categorical and criterion-based search results from a search query
CN112084405A (zh) 一种搜索方法、装置及计算机存储介质
JP2018523251A (ja) カタログ内の製品を検索するためのシステムおよび方法
US20150160847A1 (en) System and method for searching through a graphic user interface
US20160299951A1 (en) Processing a search query and retrieving targeted records from a networked database system
US9928466B1 (en) Approaches for annotating phrases in search queries
WO2023134324A1 (zh) 一种搜索请求处理方法、装置、计算机设备及存储介质
Kahya-Özyirmidokuz Analyzing unstructured Facebook social network data through web text mining: A study of online shopping firms in Turkey
US11488223B1 (en) Modification of user interface based on dynamically-ranked product attributes
CN113378058A (zh) 一种信息搜索的方法、装置、计算机设备及存储介质
US10671660B2 (en) Contextual ranking of photos and text in search

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21850390

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 16.05.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21850390

Country of ref document: EP

Kind code of ref document: A1