WO2024114681A1 - Procédé et appareil d'affichage de résultats de recherche, et dispositif informatique et support de stockage - Google Patents

Procédé et appareil d'affichage de résultats de recherche, et dispositif informatique et support de stockage Download PDF

Info

Publication number
WO2024114681A1
WO2024114681A1 PCT/CN2023/135012 CN2023135012W WO2024114681A1 WO 2024114681 A1 WO2024114681 A1 WO 2024114681A1 CN 2023135012 W CN2023135012 W CN 2023135012W WO 2024114681 A1 WO2024114681 A1 WO 2024114681A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
search
sample
key
structured
Prior art date
Application number
PCT/CN2023/135012
Other languages
English (en)
Chinese (zh)
Inventor
严林
乔超
余鑫
Original Assignee
北京字跳网络技术有限公司
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 北京字跳网络技术有限公司 filed Critical 北京字跳网络技术有限公司
Publication of WO2024114681A1 publication Critical patent/WO2024114681A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9538—Presentation of query results
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/10—Text processing
    • G06F40/194—Calculation of difference between files
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the present disclosure relates to the field of computer technology, and in particular to a search result display method, device, computer equipment and storage medium.
  • the display format of each search result in the search result display page is often the same.
  • the content displayed by some search results may not meet the user's search needs, which may cause the user to misjudge whether the search results meet their search needs.
  • the webpage content of a webpage may contain valid information that matches the user's search needs very well, but due to the limitation of the display format, part of the content of the webpage displayed in the search results does not reflect the valid information well.
  • the embodiments of the present disclosure at least provide a search result display method, apparatus, computer equipment and storage medium.
  • an embodiment of the present disclosure provides a search result display method, including:
  • search results include one of key information and structured information of the target web page content;
  • the display form corresponding to the key information includes highlighting the keywords in the key information that match the search information
  • the display form corresponding to the structured information includes structured display of field information of multiple key fields of the target web page content.
  • the search result matching the search information is determined according to the following method:
  • each search result includes the key information and the structured information
  • corresponding information with a higher correlation is selected from the key information and the structured information as a search result matching the search information.
  • the structured information is predetermined according to the following method:
  • the structured information corresponding to the target web page content is extracted; the structured information includes field information under multiple key fields indicated by the key field combination.
  • the key information is determined according to the following method:
  • Selecting corresponding information with a higher relevance from the key information and the structured information as a search result matching the search information includes:
  • information with a higher relevance score is displayed as a search result matching the search information.
  • the neural network model is trained according to the following steps:
  • sample search information Acquire sample search information, sample key information and sample structured information of webpage content corresponding to the sample search information, and a sample tag; wherein the sample tag is used to mark a first sample relevance between the sample key information and the sample search information, and a second sample relevance between the sample structured information and the sample search information;
  • a loss value of this training is determined, and parameters of the neural network model to be trained are adjusted based on the loss value.
  • the highlighting of the keywords in the key information that match the search information includes:
  • the calculated relevance between each segmentation and the search information is used to highlight the segmentation in the key information whose relevance to the search information is greater than a second set threshold. form.
  • the present disclosure also provides a search result display device, including:
  • An acquisition module configured to acquire search results matching the search information in response to receiving the search information; the search results include one of key information and structured information of the target webpage content;
  • a display module used to display the search results in a display format that matches the search results
  • the display form corresponding to the key information includes highlighting the keywords in the key information that match the search information
  • the display form corresponding to the structured information includes structured display of field information of multiple key fields of the target web page content.
  • the acquisition module is used to determine search results matching the search information in the following manner:
  • each search result includes the key information and the structured information
  • corresponding information with a higher correlation is selected from the key information and the structured information as a search result matching the search information.
  • the acquisition module is further configured to predetermine the structured information in the following manner:
  • the structured information corresponding to the target web page content is extracted; the structured information includes field information under multiple key fields indicated by the key field combination.
  • the acquisition module is further configured to determine the key information in the following manner:
  • the acquisition module when selecting corresponding information with a higher correlation from the key information and the structured information as a search result matching the search information based on a first correlation between the key information and the search information and a second correlation between the structured information and the search information, is configured to:
  • information with a higher relevance score is displayed as a search result matching the search information.
  • the acquisition module is further used to train and obtain the neural network model according to the following steps:
  • sample search information Acquire sample search information, sample key information and sample structured information of webpage content corresponding to the sample search information, and a sample tag; wherein the sample tag is used to mark a first sample relevance between the sample key information and the sample search information, and a second sample relevance between the sample structured information and the sample search information;
  • a loss value of this training is determined, and parameters of the neural network model to be trained are adjusted based on the loss value.
  • the display module when highlighting the keywords in the key information that match the search information, is used to:
  • the calculated relevance between each segmented word and the search information is used to set the segmented words in the key information whose relevance to the search information is greater than a second set threshold to be highlighted.
  • an embodiment of the present disclosure further provides a computer device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned first aspect, or any possible implementation of the first aspect are performed.
  • an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored.
  • a computer program is stored.
  • the steps of the above-mentioned first aspect, or any possible implementation of the first aspect are executed.
  • the search results matching the search information may include one of the key information and structured information of the web page content, and the search results are displayed in a display format that matches the search results.
  • the search results matching the search information may include one of the key information and structured information of the web page content, and the search results are displayed in a display format that matches the search results.
  • the key information and structured information corresponding to the web page content that best matches the search requirement can be used as the search result, and displayed in a display format corresponding to it, rather than in a single display format that limits the display of key information or structured information.
  • This method can This makes the display form of search results more flexible and makes the displayed search results better adapt to the search information, thereby better meeting the search needs of users.
  • FIG1 shows a flow chart of a search result display method provided by an embodiment of the present disclosure
  • FIG2a shows a schematic diagram of a search result display page in the search result display method provided by an embodiment of the present disclosure
  • FIG2b shows a schematic diagram of another search result display page in the search result display method provided in an embodiment of the present disclosure
  • FIG3 shows a schematic diagram of the architecture of a search result display device provided by an embodiment of the present disclosure
  • FIG4 shows a schematic diagram of the structure of a computer device provided by an embodiment of the present disclosure.
  • a and/or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone.
  • at least one herein represents any combination of at least two of any one or more of a plurality of.
  • including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.
  • a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information.
  • the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.
  • the prompt information in response to receiving an active request from the user, may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form.
  • the pop-up window may also carry a selection control for the user to choose "agree” or “disagree” to provide personal information to the electronic device.
  • the disclosed embodiment provides a search result display method, device, computer equipment and storage medium.
  • search result When it is necessary to present search results related to web page content, it can be determined which of the key information and structured information of the web page content extracted in advance best matches the search information.
  • the search result finally presented can be the key information and structured information of the web page content that best matches the search information, and the search result is displayed in a display form that matches the search result.
  • This method can well refine and display the information in the web page by displaying the field information of the key field in the web page in a structured manner, or highlighting the keywords that match the search information, thereby improving the user's browsing efficiency for the search results.
  • a kind of information that best matches the search requirement can be selected from the key information and structured information corresponding to the web page content as the search result, rather than displaying the key information or structured information in a single display form.
  • This method can make the display form of the search result more flexible, and make the displayed search results better adapt to the search information, thereby better meeting the user's search needs.
  • the execution subject of the search result display method provided in the embodiment of the present disclosure is generally a computer device with certain computing capabilities, and the computer device includes, for example: a terminal device or a server or other processing device.
  • the terminal device may be an intelligent terminal device with a display function, for example, a smart phone, a tablet computer, an intelligent wearable device, etc.
  • the search result display method may be implemented by a processor calling a computer-readable instruction stored in a memory.
  • the method includes S101 to S102, wherein:
  • search results In response to receiving search information, obtaining search results matching the search information; the search results include one of key information and structured information of target webpage content.
  • the display form corresponding to the key information includes highlighting the keywords in the key information that match the search information
  • the display form corresponding to the structured information includes highlighting the keywords in the key information that match the search information. This includes structured display of field information of multiple key fields of the target web page content.
  • the key information and structured information corresponding to the web page content can be extracted in advance; after receiving the search information, the information that best matches the search requirements can be selected from the key information and structured information corresponding to the web page content as the search result, and displayed in a corresponding display format, wherein, if the key information is selected for display, the keywords that answer the search information question can be highlighted, such as a red display, and if the structured information is selected for display, the field information of multiple key fields of the pre-extracted web page content can be displayed in a configured structured style.
  • This method can make the display format of the search results more flexible, and make the displayed search results better adapt to the search information, thereby better meeting the user's search needs.
  • the above-mentioned search information may be information entered by the user in the search bar of the search page, or may be search information determined based on recommended search terms; the target web page content is the web page content in the target web page that matches the search information, and the web page content of the target web page may include answer information corresponding to the search information.
  • the key information may include summary content such as the title and content summary of the target web page content; the structured information is used to display the information of multiple key fields in the web page content in a structured manner.
  • the key fields may include cover, title, category, author, update status, directory, etc.
  • the structured information may be displayed in the form of structured information cards, etc.
  • key information and structured information of each web page content can be determined in advance, so that when users use search information to search, they can quickly obtain key information and structured information of web page content in search results.
  • structured information corresponding to the target web page content may be extracted based on a key field combination set for the web page category to which the target web page content belongs.
  • the structured information includes field information under multiple key fields indicated by the key field combination.
  • different field information corresponds to web page content at different positions in the web page
  • different web page categories may correspond to different fields, so that different web page content can be extracted from web pages of different categories.
  • a pre-trained web page category determination model may be used to identify the web page category of each web page, thereby determining the web page category of each web page in batches.
  • the web page category determination model can be a neural network model that can be used to perform classification tasks, such as a Bidirectional Encoder Representations from Transformers (BERT) model.
  • BET Bidirectional Encoder Representations from Transformers
  • the input data of the web page category determination model can be obtained after data construction of the web page content of the target web page; the input data obtained by data construction is used to characterize the content and attribute information of each content node in the network content; the attribute information includes depth information and position information, the depth information is used to characterize the hierarchical relationship between the content node and other content nodes, and the position information is used to characterize the layout position of the content node in the web page.
  • the node tree corresponding to the target web page can be traversed based on a depth-first traversal algorithm to determine the depth information and position information corresponding to each content node; wherein the node tree is generated based on the web page code corresponding to the target web page; and according to a preset data construction order, the content of each content node, the depth information and position information corresponding to each content node are spliced to obtain the input data corresponding to the target web page.
  • the sample input data generated after data construction of the sample web page can be input into the web page category determination model to be trained to obtain the sample web page category output by the web page category recognition model, and based on the sample web page category and the web page category label corresponding to the sample web page, the loss value of this training is determined, and based on the loss value, the network parameters of the web page category determination model are adjusted until the number of parameter adjustments reaches a preset number of adjustments, and/or the network accuracy of the web page category determination model meets the preset accuracy requirements.
  • the key field combination corresponding to each web page can be configured to extract the structured information corresponding to the web page content in each web page according to the field access path corresponding to each key field in the key field combination; in order to improve the configuration efficiency of the key field combination, the same key field combination can also be configured for web pages belonging to the same web page group, and the web page group can be composed of web pages corresponding to some of the same web page addresses.
  • the target webpage content of the target webpage can be extracted according to the key field combination corresponding to the webpage category to which the webpage category of the target webpage belongs, thereby obtaining the structured information corresponding to the target webpage content.
  • A1 Extracting each text segment from the target webpage content.
  • the target webpage content may be subjected to text segmentation processing to obtain various text segments in the target webpage content.
  • text segmentation processing when performing text segmentation processing on the target webpage content, text segmentation processing may be performed according to a preset text segmentation length to obtain each text segment in the target webpage content;
  • text segmentation processing may also be performed according to target punctuation marks in the target webpage content, and the Nth target punctuation mark may be
  • the text content between the period symbol and the N+1th target punctuation symbol is regarded as a text segment, wherein N is a positive integer, and the text content before the first target punctuation symbol can be regarded as a text segment, and the target punctuation symbol can be, for example, a comma, a period, or other symbols;
  • the text segmentation processing can also be performed from the perspective of text semantics through a pre-trained text segmentation model to obtain the various text segments in the target web page content. That is, the text segmentation model divides the inseparable semantic content in the target web page content into one text segment and divides the content with different semantics into different text segments by performing semantic recognition.
  • the text can also be subjected to word segmentation processing in advance, and the input data of the text segmentation model can be the word segmentation result corresponding to the web page content, that is, a plurality of words to be combined obtained after the word segmentation processing of the web page content, and the text segmentation model can perform semantic recognition on the words to be combined, and combine the words to be combined that have the same semantics, and/or similar semantics, and/or semantically related and are located adjacent to each other in the web page content, to obtain at least one word combination, and each of the word combinations can constitute a text segment.
  • A2 Determine the relevance between each of the text segments and the search information.
  • the relevance between the text segment and the search information can be determined by calculating the Euclidean distance between the feature representation vectors of the two, the edit distance between the characters of the two, etc., or the semantic relevance of the two can be directly calculated by a semantic recognition model.
  • the embodiments of the present disclosure do not specifically limit the method of calculating the similarity.
  • A3 Select the corresponding text segment with the highest relevance as the key information; or select at least one text segment with a corresponding relevance greater than a first set threshold, and integrate the at least one text segment into the key information.
  • the corresponding text segment 5 with the highest relevance can be used as the key information; alternatively, text segments 2 and 5 whose relevances are greater than the first set threshold of 0.65 can be selected, and text segments 2 and 5 can be integrated according to the arrangement order of each text segment in the web page content, and the integrated text content can be used as the key information.
  • the target web page content may correspond to both key information and structured information. If the key information and structured information corresponding to the target web page content are displayed simultaneously in the search result display page, more content may be displayed, thereby affecting the display effect of other subsequent search results.
  • each search result includes the key information and the structured information.
  • the search result may include key information of the webpage content corresponding to the search result, and structured information of the webpage content corresponding to the search result.
  • B2 Based on the first correlation between the key information and the search information, and the second correlation between the structured information and the search information, select corresponding information with higher correlation from the key information and the structured information as the search result matching the search information.
  • the first relevance and the second relevance may be text relevance and/or semantic relevance.
  • B21 Input the search information, the key information corresponding to the search information and the structured information into a pre-trained neural network model to obtain a first correlation score output by the neural network model for characterizing a first correlation and a second correlation score output by the neural network model for characterizing a second correlation.
  • the network type of the pre-trained neural network model may be, for example, a (Bidirectional Encoder Representations from Transformer) model.
  • the corresponding structured information with a higher correlation score can be displayed as a search result matching the search information.
  • the neural network model can be trained by the following steps C1 to C3:
  • C1 Obtain sample search information, sample key information and sample structured information of the web page content corresponding to the sample search information, and a sample label; wherein the sample label is used to mark a first sample correlation between the sample key information and the sample search information, and a second sample correlation between the sample structured information and the sample search information.
  • C2 Input the sample search information, the sample key information, and the sample structured information into the neural network model to be trained, and obtain a first sample correlation score output by the neural network model to characterize a first correlation and a second sample correlation score output by the neural network model to characterize a second correlation.
  • C3 Based on the first sample-related score, the second sample-related score and the sample label, determine the loss value of this training, and adjust the parameters of the neural network model to be trained based on the loss value.
  • the loss value of this training may include a first loss value and a second loss value; wherein, the first loss value can be used to characterize the difference between the first sample correlation score and the first sample correlation degree marked by the sample label, that is, whether the first sample correlation score is the same as the marked correlation degree; the second loss value can be used to characterize the difference between the second sample correlation score and the second sample correlation degree marked by the sample label, that is, whether the second sample correlation score is the same as the marked correlation degree.
  • the first loss value and the second loss value may be weighted based on the first loss value, the second loss value and a preset weight parameter. and, based on the loss value obtained after the weighted summation, adjusting the parameters of the neural network model until the number of parameter adjustments reaches a preset number of adjustments, and/or the network accuracy of the neural network model meets the preset accuracy requirements.
  • information with higher relevance to the search information can be selected from the key information and structured information corresponding to the web page content as the search results matching the search information, so that the final displayed search results have a higher relevance to the search information, thereby improving the user's search result browsing efficiency.
  • the field information of multiple key fields of the target webpage content may be structured according to a structured display template matching the network type of the target webpage.
  • FIG2a a schematic diagram of a search result display page may be shown in FIG2a.
  • the search information input by the user is the novel search information "S
  • the target web page matching the search information is the web page corresponding to the target novel.
  • a structured display template i.e., a novel card
  • the displayed structured information includes author information, status information, introduction information, etc.
  • the recently updated novel content is also displayed in the novel card, and the user can browse the content of the corresponding novel chapter by triggering the corresponding chapter.
  • steps D1 to D2 when highlighting the keywords in the key information that match the search information, the following steps D1 to D2 may be performed:
  • D1 Perform word segmentation processing on the key information to obtain individual word segments.
  • a Chinese word segmentation model (Chinese Word Segmentation, CWS) can be used to perform word segmentation processing on the key information to obtain each word segment after word segmentation processing.
  • D2 Calculate the relevance between each segmented word and the search information, and set the segmented words in the key information whose relevance to the search information is greater than a second set threshold to be highlighted.
  • the relevance between the word segmentation and the search information may include text relevance such as Euclidean distance, edit distance, and/or semantic relevance; when determining the relevance between the word segmentation and the search information, it can be calculated according to a preset relevance calculation formula.
  • the embodiment of the present disclosure does not limit the selection of which relevance calculation formula can be based on what can be implemented in actual applications; the highlighted display form may include bolding the font, changing the font color, changing the background color of the area where the font is located, increasing the font size, and the like.
  • the word segment whose relevance to the search information is greater than a second set threshold, or the first M word segment in the relevance sorting queue can be set to be highlighted; wherein M is a preset positive integer.
  • word 1 and word 2 greater than the second set threshold value 0.55 can be used as word segments that need to be highlighted, and the display form of word 1 and word 2 can be set to the highlighted display form, so that word 1 and word 2 are highlighted when displaying the search results; alternatively, the word segments can be sorted from high to low according to the relevance, and the first three word segments "word 1", "word 2", and "word 3" in the relevance queue obtained after sorting can be used as word segments that need to be highlighted, and the display form of word 1, word 2, and word 3 can be set to the highlighted display form, so that word 1, word 2, and word 3 are highlighted when displaying the search results.
  • a schematic diagram of the search result display page can also be shown in FIG2b, where the search information input by the user is “XX Travel Notes”, and the target web pages matching the search information include web pages corresponding to the target film and television works.
  • the search results a display style matching the web page type of the target web page is used to display the key information of the target web page, and the keywords “XX Travel Notes” and “TV series” in the key information that match the search information are highlighted; wherein the key information displayed on the film and television work card includes episode information, introduction information, etc.
  • the search result display method provided by the disclosed embodiment can well refine and display the information in the web page by displaying the field information of the key fields in the web page content in a structured manner, or highlighting the key information matching the search information, thereby improving the user's browsing efficiency of the search results.
  • the key information and structured information corresponding to the web page content that best matches the search requirements can be used as the search result, rather than limiting the display of key information or structured information. This method can make the display form of the search results more flexible, and make the displayed search results better adapt to the search information, thereby better meeting the user's search needs.
  • the search results finally presented in the search results page may be the key information of the web page content in some cases, and the structured information of the web page content in other cases.
  • the displayed search results can be better adapted to the search needs, and the user can accurately predict whether the web page content meets his search needs, and then make a decision whether to further consume the web page details, thereby improving the browsing efficiency of the user's search results.
  • the embodiment of the present disclosure also provides a search result display device corresponding to the search result display method. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the above-mentioned search result display method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
  • FIG. 3 is a schematic diagram of the architecture of a search result display device provided by an embodiment of the present disclosure, wherein the device includes: an acquisition module 301 and a display module 302; wherein:
  • the acquisition module 301 is used to obtain a matching
  • the search results include one of key information and structured information of the target web page content;
  • a display module 302 configured to display the search results in a display format that matches the search results
  • the display form corresponding to the key information includes highlighting the keywords in the key information that match the search information
  • the display form corresponding to the structured information includes structured display of field information of multiple key fields of the target web page content.
  • the acquisition module 301 is used to determine search results matching the search information in the following manner:
  • each search result includes the key information and the structured information
  • corresponding information with a higher correlation is selected from the key information and the structured information as a search result matching the search information.
  • the acquisition module 301 is further configured to predetermine the structured information in the following manner:
  • the structured information corresponding to the target web page content is extracted; the structured information includes field information under multiple key fields indicated by the key field combination.
  • the acquisition module 301 is further configured to determine the key information in the following manner:
  • the acquisition module 301 when selecting corresponding information with a higher correlation from the key information and the structured information as a search result matching the search information based on a first correlation between the key information and the search information and a second correlation between the structured information and the search information, is configured to:
  • information with a higher relevance score is displayed as a search result matching the search information.
  • the acquisition module 301 is further used to train and obtain the neural network model according to the following steps:
  • sample search information Acquire sample search information, sample key information and sample structured information of webpage content corresponding to the sample search information, and a sample tag; wherein the sample tag is used to mark a first sample relevance between the sample key information and the sample search information, and a second sample relevance between the sample structured information and the sample search information;
  • a loss value of this training is determined, and parameters of the neural network model to be trained are adjusted based on the loss value.
  • the display module 302 in the key information When the keywords matching the above search information are highlighted, it is used to:
  • the calculated relevance between each segmented word and the search information is used to set the segmented words in the key information whose relevance to the search information is greater than a second set threshold to be highlighted.
  • the search result matching the search information may include one of the key information and structured information of the webpage content, and the search result is displayed in a display form matching the search result (the key information and the structured information correspond to different display forms respectively).
  • This method can make the search results better adapt to different search information, and can enable users to accurately predict whether the webpage content meets their search needs, and then make a decision whether to further consume the webpage details, thereby improving the browsing efficiency of the user's search results.
  • the embodiment of the present disclosure also provides a computer device.
  • FIG. 4 it is a schematic diagram of the structure of the computer device 400 provided in the embodiment of the present disclosure, including a processor 401, a memory 402, and a bus 403.
  • the memory 402 is used to store execution instructions, including a memory 4021 and an external memory 4022; the memory 4021 here is also called an internal memory, which is used to temporarily store the calculation data in the processor 401, and the data exchanged with the external memory 4022 such as a hard disk.
  • the processor 401 exchanges data with the external memory 4022 through the memory 4021.
  • the processor 401 communicates with the memory 402 through the bus 403, so that the processor 401 executes the following instructions:
  • the search results include one of key information and structured information of the target webpage content;
  • the display form corresponding to the key information includes highlighting the keywords in the key information that match the search information
  • the display form corresponding to the structured information includes structured display of field information of multiple key fields of the target web page content.
  • the search result matching the search information is determined according to the following method:
  • each search result includes the key information and the structured information
  • corresponding information with a higher correlation is selected from the key information and the structured information as a search result matching the search information.
  • the structured information is predetermined according to the following method:
  • the structured information corresponding to the target web page content is extracted; the structured information includes field information under multiple key fields indicated by the key field combination.
  • the key information is determined according to the following method:
  • the instructions of the processor 401 are based on the first correlation between the key information and the search information, and the correlation between the structured information and the search information.
  • the method further comprises: determining a second correlation between the key information and the structured information, selecting corresponding information with a higher correlation from the key information and the structured information as a search result matching the search information, including:
  • information with a higher relevance score is displayed as a search result matching the search information.
  • the neural network model is trained according to the following steps:
  • sample search information Acquire sample search information, sample key information and sample structured information of webpage content corresponding to the sample search information, and a sample tag; wherein the sample tag is used to mark a first sample relevance between the sample key information and the sample search information, and a second sample relevance between the sample structured information and the sample search information;
  • a loss value of this training is determined, and parameters of the neural network model to be trained are adjusted based on the loss value.
  • highlighting the keywords in the key information that match the search information includes:
  • the calculated relevance between each segmented word and the search information is used to set the segmented words in the key information whose relevance to the search information is greater than a second set threshold to be highlighted.
  • the present disclosure also provides a computer-readable storage medium on which a computer program is stored.
  • a computer program When the computer program is executed by a processor, the steps of the search result display method described in the above method embodiment are executed.
  • the storage medium may be a volatile or non-volatile computer-readable storage medium.
  • the present disclosure also provides a computer program product that carries a program code.
  • the program code includes instructions that can be used to execute the steps of the search result display method described in the above method embodiment. For details, please refer to the above method embodiment, which will not be repeated here.
  • the computer program product may be implemented in hardware, software or a combination thereof.
  • the computer program product is embodied as a computer storage medium.
  • the computer program product is embodied as a software product, such as a software development kit (SDK).
  • SDK software development kit
  • the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., may be located in one place.
  • the purpose of the solution of this embodiment can be achieved by selecting some or all of the units according to actual needs.
  • each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that can be executed by a processor.
  • the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente divulgation concerne un procédé et un appareil d'affichage de résultats de recherche, ainsi qu'un dispositif informatique et un support de stockage. Le procédé consiste à : en réponse à la réception d'informations de recherche, acquérir un résultat de recherche correspondant aux informations de recherche, le résultat de recherche comprenant des informations clés ou des informations structurées du contenu de la page Web cible ; et afficher le résultat de recherche dans une forme d'affichage correspondant au résultat de recherche, la forme d'affichage correspondant aux informations clés comprenant l'affichage en surbrillance d'un mot-clé correspondant aux informations de recherche dans les informations clés, et la forme d'affichage correspondant aux informations structurées comprenant l'affichage structuré des informations de champ d'une pluralité de champs clés du contenu de la page Web cible.
PCT/CN2023/135012 2022-12-01 2023-11-29 Procédé et appareil d'affichage de résultats de recherche, et dispositif informatique et support de stockage WO2024114681A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211531138.5 2022-12-01
CN202211531138.5A CN115730158A (zh) 2022-12-01 2022-12-01 一种搜索结果展示方法、装置、计算机设备及存储介质

Publications (1)

Publication Number Publication Date
WO2024114681A1 true WO2024114681A1 (fr) 2024-06-06

Family

ID=85300296

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/135012 WO2024114681A1 (fr) 2022-12-01 2023-11-29 Procédé et appareil d'affichage de résultats de recherche, et dispositif informatique et support de stockage

Country Status (2)

Country Link
CN (1) CN115730158A (fr)
WO (1) WO2024114681A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115730158A (zh) * 2022-12-01 2023-03-03 北京字跳网络技术有限公司 一种搜索结果展示方法、装置、计算机设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105138535A (zh) * 2015-06-30 2015-12-09 百度在线网络技术（北京）有限公司 一种搜索结果的展示方法及装置
CN105786847A (zh) * 2014-12-22 2016-07-20 北京奇虎科技有限公司 一种电商网站中商品网页的结构化摘要的显示方法和系统
CN111310078A (zh) * 2020-02-10 2020-06-19 北京字节跳动网络技术有限公司 一种搜索结果展示策略的确定方法及装置
US11048711B1 (en) * 2018-12-19 2021-06-29 First American Financial Corporation System and method for automated classification of structured property description extracted from data source using numeric representation and keyword search
CN115730158A (zh) * 2022-12-01 2023-03-03 北京字跳网络技术有限公司 一种搜索结果展示方法、装置、计算机设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105786847A (zh) * 2014-12-22 2016-07-20 北京奇虎科技有限公司 一种电商网站中商品网页的结构化摘要的显示方法和系统
CN105138535A (zh) * 2015-06-30 2015-12-09 百度在线网络技术（北京）有限公司 一种搜索结果的展示方法及装置
US11048711B1 (en) * 2018-12-19 2021-06-29 First American Financial Corporation System and method for automated classification of structured property description extracted from data source using numeric representation and keyword search
CN111310078A (zh) * 2020-02-10 2020-06-19 北京字节跳动网络技术有限公司 一种搜索结果展示策略的确定方法及装置
CN115730158A (zh) * 2022-12-01 2023-03-03 北京字跳网络技术有限公司 一种搜索结果展示方法、装置、计算机设备及存储介质

Also Published As

Publication number Publication date
CN115730158A (zh) 2023-03-03

Similar Documents

Publication Publication Date Title
WO2022116537A1 (fr) Procédé et appareil de recommandation de nouvelles, ainsi que dispositif électronique et support de stockage
WO2020108608A1 (fr) Procédé de traitement de résultat de recherche, dispositif, terminal, dispositif électronique et support de stockage
US11100124B2 (en) Systems and methods for similarity and context measures for trademark and service mark analysis and repository searches
CN107704512B (zh) 基于社交数据的金融产品推荐方法、电子装置及介质
US9323741B2 (en) System and method for searching functions having symbols
WO2021068932A1 (fr) Procédé basé sur un livre électronique pour présenter des informations associées à une entité
CN111459977B (zh) 自然语言查询的转换
CN111324771B (zh) 视频标签的确定方法、装置、电子设备及存储介质
WO2020233131A1 (fr) Procédé et appareil de traitement de question-réponse, dispositif informatique et support de stockage
CN110222045A (zh) 一种数据报表获取方法、装置及计算机设备、存储介质
CN105843796A (zh) 一种微博情感倾向分析方法及装置
WO2024114681A1 (fr) Procédé et appareil d'affichage de résultats de recherche, et dispositif informatique et support de stockage
CN112989208B (zh) 一种信息推荐方法、装置、电子设备及存储介质
CN112052424B (zh) 一种内容审核方法及装置
CN112132238A (zh) 一种识别隐私数据的方法、装置、设备和可读介质
WO2021012958A1 (fr) Procédé de détection de texte original, appareil, dispositif et support de stockage lisible par ordinateur
CN117743390B (zh) 针对金融信息的查询方法和系统、存储介质
CN111814481A (zh) 购物意图识别方法、装置、终端设备及存储介质
Zanibbi et al. Math search for the masses: Multimodal search interfaces and appearance-based retrieval
CN104881446A (zh) 搜索方法及装置
CN112527963B (zh) 基于词典的多标签情感分类方法及装置、设备、存储介质
CN110795942A (zh) 基于语义识别的关键词确定方法、装置和存储介质
Brek et al. Enhancing information extraction process in job recommendation using semantic technology
CN112685544A (zh) 电信信息的查询方法、装置、设备和介质
CN110688559A (zh) 一种检索方法及装置

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: 23896825

Country of ref document: EP

Kind code of ref document: A1