Retrieval method and device, terminal, storage medium and computer program product

Abstract

The embodiment of the invention provides a retrieval method and device, a terminal, a storage medium and a computer program product. The method comprises the steps that scene mode features corresponding to a target object retrieved by a user are determined through keyword data; determining attention scene features of the user according to the browsing behavior data of the user; and comparing the scene mode features with the attention scene features to determine the similarity between each scene mode feature and the attention scene features, and finally determining the display mode of the target object in the retrieval result according to the similarity. According to the scheme, the display mode of the target object retrieved in the retrieval result can be determined according to the retrieval object of the user and the browsing behavior data of the user. According to the method, the demand of the user can be dynamically met according to the behavior preference of the user, the display mode of the retrieval object in the retrieval result is determined, and the problem that the display requirement of the user for the personalized effect and the precision of the retrieval result is difficult to meet since the retrieval object is displayed only according to the matching degree of the retrieval result and the to-be-retrieved information is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, specifically to a retrieval method, device, terminal, storage medium, and computer program product. Background Technology

[0002] Currently, after users enter their search query into a business website or app's search window, the platform searches its database for the input and displays the results on the website or app's results page based on the match between the search results and the input. However, users expect both comprehensive and accurate results, as well as a reasonable display format and precise recommendations.

[0003] Therefore, simply displaying search results based on their matching degree with the search input information is insufficient to meet users' requirements for personalized and accurate search results. Summary of the Invention

[0004] At least one embodiment of this application provides a retrieval method, apparatus, terminal, storage medium, and computer program product to solve the problem in the prior art that it is difficult to associate user identifiers on the service side and the network side.

[0005] To solve the above-mentioned technical problems, this application is implemented as follows:

[0006] In a first aspect, embodiments of this application provide a retrieval method, including:

[0007] Identify multiple target objects in the database that correspond to the object to be retrieved;

[0008] The scene pattern features corresponding to each target object are determined based on keyword data; the keyword data includes comment data and attribute data.

[0009] The user's attention scenario characteristics are determined based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data.

[0010] The similarity between the scene features of interest and the scene pattern features is obtained by comparing each scene pattern feature with the scene features of interest.

[0011] The target display method is determined based on the similarity, and the target object is displayed in the search results according to the target display method.

[0012] Optionally, the scene pattern features corresponding to each target object are determined based on keyword data, including:

[0013] Construct a target keyword tree based on the keyword data;

[0014] Based on the target keyword tree, determine the scene pattern features corresponding to each target object.

[0015] Optionally, based on the target keyword tree, the scene pattern features corresponding to each target object are determined, including:

[0016] Determine the target keywords for each target object;

[0017] Calculate the first weight value of each branch corresponding to the target keyword in the keyword tree;

[0018] The scene pattern features are determined based on the first keyword data corresponding to the branch with the largest first weight value.

[0019] Optionally, calculating the first weight value of each branch corresponding to the target keyword in the keyword tree includes:

[0020] Identify all first branches in the keyword tree corresponding to the target keyword;

[0021] The sum of the weights of all keywords on the first branch is determined to be the first weight value;

[0022] The weight of a keyword is related to its frequency of occurrence in the keyword data and its importance.

[0023] Optionally, constructing a target keyword tree based on the keyword data includes:

[0024] The target keyword tree is constructed by taking the nodes of the target keyword tree as the target keywords and taking the branches of the keywords as the first connection relationship between the target keywords.

[0025] The first connection relationship is determined based on the order in which the target keyword appears in the keyword data.

[0026] Optionally, user attention scenario characteristics can be determined based on behavioral keyword data, including:

[0027] Based on the behavioral keyword data, construct a tree of keywords of interest;

[0028] Based on the keyword tree of interest, the user's interest scenario characteristics are determined.

[0029] Optionally, the user's attention scenario characteristics can be determined based on the keyword tree, including:

[0030] Determine the keywords of interest corresponding to the browsing behavior data;

[0031] Calculate the second weight value of the keyword in each branch of the keyword tree;

[0032] The features of the focus scene are determined based on the second keyword data corresponding to the branch with the largest second weight value.

[0033] Optionally, calculating the second weight value of each branch corresponding to the keyword in the keyword tree includes:

[0034] Identify all second branches corresponding to the keywords of interest in the keyword tree;

[0035] The sum of the weights of all keywords on the second branch is determined to be the second weight value;

[0036] The weight of the keyword is related to the frequency of its appearance in the browsing behavior data and the importance of the keyword.

[0037] Optionally, based on the behavioral keyword data, a tree of attention keywords is constructed, including:

[0038] Based on the behavioral keyword data, determine the keywords of interest;

[0039] The tree of keywords of interest is constructed by using the keywords of interest as nodes in the tree of keywords of interest and the second connection relationships between the trees of keywords of interest as branches of the keywords of interest; wherein, the behavioral keyword data is determined by the user's historical browsing behavior data;

[0040] The second connection relationship is determined based on the importance of the keywords of interest, which is related to the frequency, duration of stay, and number of searches of the keywords of interest in the historical browsing behavior data.

[0041] Secondly, embodiments of this application provide a retrieval device, including:

[0042] The first determination module is used to determine multiple target objects in the database that correspond to the object to be retrieved.

[0043] The second determining module is used to determine the scene pattern features corresponding to each target object based on keyword data; the keyword data includes comment data and attribute data.

[0044] The third determining module is used to determine the user's attention scene characteristics based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data.

[0045] The comparison module is used to compare the features of the scene of interest with the features of the scene pattern to obtain the similarity between each feature of the scene pattern and the features of the scene of interest.

[0046] The fourth determining module is used to determine the target display method based on the similarity, and to display the target object in the search results according to the target display method.

[0047] Thirdly, embodiments of this application provide a terminal, including: a processor;

[0048] The processor is used to determine multiple target objects in the database that correspond to the object to be retrieved;

[0049] The scene pattern features corresponding to each target object are determined based on keyword data; the keyword data includes comment data and attribute data.

[0050] The user's attention scenario characteristics are determined based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data.

[0051] The similarity between the scene features of interest and the scene pattern features is obtained by comparing each scene pattern feature with the scene features of interest.

[0052] The target display method is determined based on the similarity, and the target object is displayed in the search results according to the target display method.

[0053] Fourthly, embodiments of this application provide a terminal, including: a transceiver, a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the method described in the first aspect.

[0054] Fifthly, embodiments of this application provide a computer-readable storage medium storing a program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0055] Sixthly, embodiments of this application provide a computer program product including computer instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0056] Compared with existing technologies, the retrieval method, apparatus, terminal, storage medium, and computer program product provided in this application determine the scene pattern features corresponding to the user's target object through keyword data; determine the user's attention scene features based on the user's browsing behavior data; then compare the scene pattern features with the attention scene features to determine the similarity between each scene pattern feature and the attention scene features, and finally determine the display method of the target object in the retrieval results based on the similarity. The solution of this application can determine the display method of the target object in the retrieval results based on the user's retrieval object and the user's browsing behavior data. It can dynamically adapt to the user's needs based on the user's behavioral preferences and determine the display method of the retrieval object in the retrieval results, solving the problem that existing technologies only display based on the matching degree between the retrieval results and the retrieval input information, which is difficult to meet the user's requirements for personalized effects and accuracy in displaying retrieval results. Attached Figure Description

[0057] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0058] Figure 1 This is a schematic diagram illustrating the steps of the retrieval method according to an embodiment of this application;

[0059] Figure 2 This is a schematic diagram of the target keyword tree structure in an embodiment of this application;

[0060] Figure 3 This is a schematic diagram of the structure of the keyword tree in an embodiment of this application;

[0061] Figure 4 This is a schematic diagram of the retrieval device according to an embodiment of this application;

[0062] Figure 5 This is a structural diagram of the terminal according to an embodiment of this application;

[0063] Figure 6 This is a schematic diagram of the structure of a terminal according to an embodiment of this application. Detailed Implementation

[0064] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, without limiting the number of objects; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, "A or B" covers three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0065] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc.; an indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0066] It is worth noting that the technologies described in this application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA), or other systems. The terms "system" and "network" in this application are often used interchangeably, and the described technologies can be used in the systems and radio technologies mentioned above, as well as in other systems and radio technologies. The following description describes New Radio (NR) systems for illustrative purposes, and the term NR is used in most of the following description; however, these technologies can also be applied to systems other than NR systems, such as 6th Generation (6G) communication systems.

[0067] To enable those skilled in the art to better understand the embodiments of this application, the following description is provided first:

[0068] With the rapid development of computer vision and internet technology, various websites and terminals have brought convenience to people's lives and entertainment. Quickly and accurately matching the user's expected search results from massive amounts of data, products, information, and business based on user input has become a major research direction for improving user experience.

[0069] In existing technologies, after people enter the content of the object to be searched through the search window of a business website or APP, the business platform or APP platform will search the target database based on the search input information, and display the search results to the user on the results display page of the business website or APP according to the matching of the search results with the search input information.

[0070] This technology only searches and matches results against a target database based on user input, then displays the results according to the degree of match between the search results and the input. While this provides a relatively comprehensive overview of the user's desired search, users expect both a comprehensive and accurate display of the searched content and a reasonable presentation and precise recommendations of the matched items. Current technology fails to consider these user needs, resulting in search results that do not meet user requirements and a degraded user experience.

[0071] For example, in the China Mobile app, users can search for information by entering "ringback tone" in the search bar on the "Search Services" page. The search results are generally displayed in descending order of the subscription volume of the corresponding application or service.

[0072] As described in the background section, in the prior art, the search is only performed in the target database based on the search input information entered by the user, and then the results are displayed according to the degree of matching between the search results and the search input information. This makes it difficult to meet the user's requirements for personalized and accurate display of search results. In order to solve the above problems, the embodiments of this application provide a search method, device, terminal, storage medium and computer program product, which can reduce or avoid the occurrence of the above situations and improve the user experience.

[0073] This application provides a relay link connection control method and apparatus. The method and apparatus are based on the same concept, and since the principles by which they solve the problem are similar, their implementations can be referred to interchangeably; repeated details will not be repeated.

[0074] like Figure 1As shown in the embodiment of this application, a retrieval method includes the following steps:

[0075] Step 101: Identify multiple target objects in the database that correspond to the object to be retrieved;

[0076] Step 102: Determine the scene pattern features corresponding to each target object based on the keyword data; the keyword data includes comment data and attribute data;

[0077] Step 103: Determine the user's attention scenario characteristics based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data;

[0078] Step 104: Compare the features of the scene of interest with the features of the scene pattern to obtain the similarity between each feature of the scene pattern and the features of the scene of interest;

[0079] Step 105: Determine the target display method based on the similarity, and display the target object in the search results according to the target display method.

[0080] Optionally, the database is a pre-built database;

[0081] The objects mentioned include business, activities, products, lifestyle, government and enterprises, etc.

[0082] For example, when a user searches for a service or business, such as "ringback tone", the database is a business database of service information provided by the operator's official APP;

[0083] Match content related to the object retrieved by the user in the business database.

[0084] In this embodiment of the application, the display method of each target object in the search results is initially determined by the relevance between the object to be searched and the target objects in the database.

[0085] For example, taking the ringback tone object as an example, after a user searches for "ringback tone", they get several different objects to be displayed, including: Douyin ringback tone, video ringback tone membership, ringback tone carnival, video ringback tone basic membership package, AI ringback tone, Douyin ringback tone, etc.; the display method of the object to be displayed is determined according to the relevance between the object to be displayed and "ringback tone".

[0086] Optionally, the keyword data is obtained by acquiring the object's comment data and attribute data.

[0087] It should be noted that the attribute data refers to the basic information of the object, including name, type, and description, which can be obtained through the object details.

[0088] Optionally, the target display method is to display the target objects one by one according to the similarity from large to small.

[0089] Optionally, obtaining the keyword data by acquiring the object's comment data and details page includes:

[0090] Retrieve the object's comment data and the object's data from the details page;

[0091] The keyword data is obtained by identifying the object data;

[0092] The keyword data includes at least one of the following: image data, text data, and video data.

[0093] For example, object review data includes user comments after ordering, purchasing, or using the object, as well as sharing summary data when users share the object. Review data can be collected periodically via telephone, voice, or an app. Attribute data is an important basis for determining object scenario patterns. For example, images include main function descriptions, videos include audio introductions and trial demonstration images, and text includes object rule parameters, etc.

[0094] Optionally, retrieve the object's comment data and the object data from the details page, including at least one of the following:

[0095] Image recognition algorithms are used to identify image data and obtain keywords for detailed images;

[0096] Image and semantic recognition are performed on the video data to obtain video detail keywords;

[0097] The text data is extracted using a summary generation algorithm to obtain summary keywords.

[0098] Optionally, the object data is identified to obtain the keyword data, including:

[0099] The object data is segmented to obtain a group of keywords;

[0100] The keyword data is determined based on the keyword group.

[0101] Optionally, determining the keyword data based on the keyword group includes:

[0102] An embedding layer is used to convert the keyword groups into low-dimensional word vectors;

[0103] The word vectors include image keyword vectors, video keyword vectors, and abstract keyword vectors.

[0104] The retrieval method provided in this application determines the scene pattern features corresponding to the target object searched by the user through keyword data; determines the user's attention scene features based on the user's browsing behavior data; then compares the scene pattern features with the attention scene features to determine the similarity between each scene pattern feature and the attention scene features, and finally determines the display method of the target object in the search results based on the similarity. This application's solution can determine the display method of the target object in the search results based on the user's search object and browsing behavior data. It can dynamically adapt to the user's needs based on their behavioral preferences and determine the display method of the search object in the search results, solving the problem that the existing technology only displays results based on the matching degree between the search results and the search input information, which is difficult to meet the user's requirements for personalized and accurate display of search results.

[0105] Optionally, the scene pattern features corresponding to each target object are determined based on keyword data, including:

[0106] Construct a target keyword tree based on the keyword data;

[0107] Based on the target keyword tree, determine the scene pattern features corresponding to each target object.

[0108] Optionally, the method further includes:

[0109] The importance of the image keyword vector, the importance of the video keyword vector, and the importance of the comment summary keyword vector are determined.

[0110] Specifically, cosine distance is used to calculate the similarity between vectors:

[0111]

[0112] in,( ) represents the image keyword vector, ( ) represents the video keyword vector, ( Abstract keyword vector.

[0113] Furthermore, the importance of each mode is calculated:

[0114]

[0115] in, , ;

[0116] Similarly, calculation and .

[0117] Then, importance normalization is performed:

[0118] ;

[0119] ;

[0120] .

[0121] It should be noted that the keyword display type decision model is constructed as follows:

[0122] 1. Image priority criteria:

[0123] ; (suggestion );

[0124] 2. Video priority criteria:

[0125] ;

[0126] 3. Abstract priority criteria:

[0127] ;

[0128] When a single priority condition is not met, a combined display is used: Display weight = ;

[0129] By calculating the importance above, we can determine the keyword display type preference. When the importance of the keyword vector representation in the detail image is greater than that of other types of keyword vector representations, it means that the object is suitable to be displayed as an image.

[0130] Optionally, based on the target keyword tree, the scene pattern features corresponding to each target object are determined, including:

[0131] Determine the target keywords for each target object;

[0132] Calculate the first weight value of each branch corresponding to the target keyword in the keyword tree;

[0133] The scene pattern features are determined based on the first keyword data corresponding to the branch with the largest first weight value.

[0134] Optionally, calculating the first weight value of each branch corresponding to the target keyword in the keyword tree includes:

[0135] Identify all first branches in the keyword tree corresponding to the target keyword;

[0136] The sum of the weights of all keywords on the first branch is determined to be the first weight value;

[0137] The weight of a keyword is related to its frequency of occurrence in the keyword data and its importance.

[0138] Optionally, constructing a target keyword tree based on the keyword data includes:

[0139] The target keyword tree is constructed by taking the nodes of the target keyword tree as the target keywords and taking the branches of the keywords as the first connection relationship between the target keywords.

[0140] The first connection relationship is determined based on the order in which the target keyword appears in the keyword data.

[0141] Optionally, the target keyword tree is constructed based on the pre-determined image keyword vector, video keyword vector, and comment summary keyword vector.

[0142] For example, the image keyword vector, the video keyword vector, and the comment summary keyword vector are used as nodes of the target keyword tree. The connection relationship of the keyword tree is determined by the weight value of the determined keyword and the frequency of the occurrence of the keyword branch, thereby constructing the target keyword tree.

[0143] like Figure 2 As shown, keywords 1 to 5 may appear on the details page of the same object. Keywords 1 to 5 can be different types of keyword vectors; for example, keyword 1 might be a keyword vector for the details image, and keyword 2 for the details video. The order of keywords 1 to 5 is determined by their distribution on the details page; for example, the order of keywords in the keyword tree is determined by their position on the details page. Furthermore, the weight of each branch, such as keywords 1 to 5, is determined by the frequency of the corresponding keyword in each object and the importance of the keyword within the branch. Specifically, the branch weight is determined by the frequency of occurrence, and the weight of each keyword within the branch is determined by its importance. The branch weight can measure the importance of the keyword branch within the object, and the keyword weight can measure the preference for the display type of the keyword on the details page, such as text, video, or image.

[0144] The retrieval method of this application embodiment can randomly arrange and combine keywords extracted from the object details page to obtain multiple different keyword combinations by constructing a keyword tree. Then, it searches in the keyword tree and selects the keyword combination vector corresponding to the largest branch weight as the object scene mode feature, and uses its weight as the object scene mode weight.

[0145] Optionally, user attention scenario characteristics can be determined based on behavioral keyword data, including:

[0146] Based on the behavioral keyword data, construct a tree of keywords of interest;

[0147] Based on the keyword tree of interest, the user's interest scenario characteristics are determined.

[0148] Optionally, based on the behavioral keyword data, a tree of attention keywords is constructed, including:

[0149] Based on the behavioral keyword data, determine the keywords of interest;

[0150] The tree of keywords of interest is constructed by using the keywords of interest as nodes in the tree of keywords of interest and the second connection relationships between the trees of keywords of interest as branches of the keywords of interest; wherein, the behavioral keyword data is determined by the user's historical browsing behavior data;

[0151] The second connection relationship is determined based on the importance of the keywords of interest, which is related to the frequency, duration of stay, and number of searches of the keywords of interest in the historical browsing behavior data.

[0152] Optionally, the user browsing behavior data is a user browsing behavior log.

[0153] It should be noted that the user browsing behavior data needs to be processed.

[0154] Specifically, first, user browsing behavior data is collected, invalid and abnormal data is removed, word segmentation is performed, and keywords of interest are extracted.

[0155] Next, the extracted keywords are traversed through the constructed keyword tree to filter out the behavioral branches that include the keywords.

[0156] Optionally, the nodes of the behavior tree are the keywords of interest extracted from the historical browsing behavior logs of all users, and the connection relationship between the nodes is determined by the importance of each keyword (which can be determined by the frequency of keyword occurrence, the market where the user stayed, the number of searches, etc.).

[0157] For example, such as Figure 3 As shown, keywords 1, 4, 7, and 9 are keywords extracted from a user's browsing behavior logs within a certain time period. Keyword 1 appeared multiple times during this period, therefore its importance is high, and its node level in the constructed behavior tree is low. Importance is used as the node's attention weight.

[0158] Finally, the weights of each behavior branch are calculated, and the behavior branch with the largest weight sum is selected as the feature of the focus scene.

[0159] Optionally, the user's attention scenario characteristics can be determined based on the keyword tree, including:

[0160] Determine the keywords of interest corresponding to the browsing behavior data;

[0161] Calculate the second weight value of the keyword in each branch of the keyword tree;

[0162] The features of the focus scene are determined based on the second keyword data corresponding to the branch with the largest second weight value.

[0163] Optionally, calculating the second weight value of each branch corresponding to the keyword in the keyword tree includes:

[0164] Identify all second branches corresponding to the keywords of interest in the keyword tree;

[0165] The sum of the weights of all keywords on the second branch is determined to be the second weight value;

[0166] The weight of the keyword is related to the frequency of its appearance in the browsing behavior data and the importance of the keyword.

[0167] Optionally, based on the behavioral keyword data, a tree of attention keywords is constructed, including:

[0168] Based on the behavioral keyword data, determine the keywords of interest;

[0169] The tree of keywords of interest is constructed by using the keywords of interest as nodes in the tree of keywords of interest and the second connection relationships between the trees of keywords of interest as branches of the keywords of interest; wherein, the behavioral keyword data is determined by the user's historical browsing behavior data;

[0170] The second connection relationship is determined based on the importance of the keywords of interest, which is related to the frequency, duration of stay, and number of searches of the keywords of interest in the historical browsing behavior data.

[0171] Optionally, the target display method includes:

[0172] The display order and display method of the target objects.

[0173] Specifically, the target objects are displayed in descending order of similarity.

[0174] Furthermore, the display method of the target object is determined based on the importance of various types of behavioral keywords.

[0175] For example, if the data of the image class is of high importance, the target object is displayed in the form of an image.

[0176] like Figure 4 As shown in the embodiments of this application, a retrieval device is also provided, including:

[0177] The first determining module 401 is used to determine multiple target objects in the database that correspond to the object to be retrieved.

[0178] The second determining module 402 is used to determine the scene mode features corresponding to each target object based on keyword data; the keyword data includes comment data and attribute data.

[0179] The third determining module 403 is used to determine the user's attention scene characteristics based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data.

[0180] The comparison module 404 is used to compare the features of the scene of interest with the features of the scene pattern to obtain the similarity between each feature of the scene pattern and the features of the scene of interest.

[0181] The fourth determining module 405 is used to determine the target display method based on the similarity, and display the target object in the search results according to the target display method.

[0182] The retrieval device provided in this application determines the scene pattern features corresponding to the target object searched by the user through keyword data; determines the user's attention scene features based on the user's browsing behavior data; then compares the scene pattern features with the attention scene features to determine the similarity between each scene pattern feature and the attention scene features, and finally determines the display method of the target object in the search results based on the similarity. The solution of this application can determine the display method of the target object in the search results based on the user's search object and the user's browsing behavior data. It can dynamically adapt to the user's needs based on the user's behavioral preferences and determine the display method of the search object in the search results, solving the problem that the existing technology only displays based on the matching degree between the search results and the search input information, which is difficult to meet the user's requirements for personalized effects and accuracy in displaying search results.

[0183] like Figure 5 As shown, this application embodiment also provides a terminal, including: a processor 501;

[0184] The processor 501 is used to determine multiple target objects in the database that correspond to the object to be retrieved.

[0185] The scene pattern features corresponding to each target object are determined based on keyword data; the keyword data includes comment data and attribute data.

[0186] The user's attention scenario characteristics are determined based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data.

[0187] The similarity between the scene features of interest and the scene pattern features is obtained by comparing each scene pattern feature with the scene features of interest.

[0188] The target display method is determined based on the similarity, and the target object is displayed in the search results according to the target display method.

[0189] The terminal provided in this application determines the scene pattern features corresponding to the user's target object through keyword data; determines the user's attention scene features based on the user's browsing behavior data; then compares the scene pattern features with the attention scene features to determine the similarity between each scene pattern feature and the attention scene features, and finally determines the display method of the target object in the search results based on the similarity. This application's solution can determine the display method of the target object in the search results based on the user's search object and browsing behavior data. It can dynamically adapt to the user's needs based on their behavioral preferences and determine the display method of the search object in the search results, solving the problem that the existing technology only displays results based on the matching degree between the search results and the search input information, which is difficult to meet the user's requirements for personalized and accurate display of search results.

[0190] Another embodiment of this application provides a terminal, such as... Figure 6 As shown, it includes a transceiver 610, a processor 600, a memory 620, and a program or instructions stored in the memory 620 and executable on the processor 600; when the processor 600 executes the program or instructions, it implements the various processes of the above-described terminal-side method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0191] The transceiver 610 is used to receive and send data under the control of the processor 600.

[0192] Among them, Figure 6In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 600 and memory represented by memory 620 together. The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. Transceiver 610 can be multiple elements, including transmitters and receivers, providing a unit for communicating with various other devices over a transmission medium. For different user equipment, user interface 630 can also be an interface capable of connecting external or internal devices, including but not limited to keypads, displays, speakers, microphones, joysticks, etc.

[0193] The processor 600 is responsible for managing the bus architecture and general processing, while the memory 620 can store the data used by the processor 600 during operation.

[0194] This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described retrieval method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0195] This application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the various processes of the above-described retrieval method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0196] It should be noted that the collection, gathering, updating, analysis, processing, use, transmission, and storage of user personal information involved in this disclosed technical solution all comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken to prevent unauthorized access to user personal information data and to safeguard user personal information security and network security.

[0197] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0198] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0199] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A retrieval method, characterized in that, include: Identify multiple target objects in the database that correspond to the object to be retrieved; The scene pattern features corresponding to each target object are determined based on keyword data; the keyword data includes comment data and attribute data. The user's attention scenario characteristics are determined based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data. The similarity between the scene features of interest and the scene pattern features is obtained by comparing each scene pattern feature with the scene features of interest. The target display method is determined based on the similarity, and the target object is displayed in the search results according to the target display method.

2. The method according to claim 1, characterized in that, Based on keyword data, the scene pattern features corresponding to each target object are determined, including: Construct a target keyword tree based on the keyword data; Based on the target keyword tree, determine the scene pattern features corresponding to each target object.

3. The method according to claim 2, characterized in that, Based on the target keyword tree, determine the scene pattern features corresponding to each target object, including: Determine the target keywords for each target object; Calculate the first weight value of each branch corresponding to the target keyword in the keyword tree; The scene pattern features are determined based on the first keyword data corresponding to the branch with the largest first weight value.

4. The method according to claim 3, characterized in that, Calculating the first weight value of each branch corresponding to the target keyword in the keyword tree includes: Identify all first branches in the keyword tree corresponding to the target keyword; The sum of the weights of all keywords on the first branch is determined to be the first weight value; The weight of a keyword is related to its frequency of occurrence in the keyword data and its importance.

5. The method according to claim 2, characterized in that, Constructing a target keyword tree based on the keyword data includes: The target keyword tree is constructed by taking the nodes of the target keyword tree as the target keywords and taking the branches of the keywords as the first connection relationship between the target keywords. The first connection relationship is determined based on the order in which the target keyword appears in the keyword data.

6. The method according to claim 1, characterized in that, Based on behavioral keyword data, we can determine the characteristics of users' attention scenarios, including: Based on the behavioral keyword data, construct a tree of keywords of interest; Based on the keyword tree of interest, the user's interest scenario characteristics are determined.

7. The method according to claim 6, characterized in that, Based on the keyword tree, the characteristics of users' attention scenarios are determined, including: Determine the keywords of interest corresponding to the browsing behavior data; Calculate the second weight value of the keyword in each branch of the keyword tree; The features of the focus scene are determined based on the second keyword data corresponding to the branch with the largest second weight value.

8. The method according to claim 7, characterized in that, Calculating the second weight value of the keyword in each branch of the keyword tree includes: Identify all second branches corresponding to the keywords of interest in the keyword tree; The sum of the weights of all keywords on the second branch is determined to be the second weight value; The weight of the keyword is related to the frequency of its appearance in the browsing behavior data and the importance of the keyword.

9. The method according to claim 6, characterized in that, Based on the behavioral keyword data, a tree of keywords of interest is constructed, including: Based on the behavioral keyword data, determine the keywords of interest; The tree of keywords of interest is constructed by using the keywords of interest as nodes in the tree of keywords of interest and the second connection relationships between the trees of keywords of interest as branches of the keywords of interest; wherein, the behavioral keyword data is determined by the user's historical browsing behavior data; The second connection relationship is determined based on the importance of the keywords of interest, which is related to the frequency, duration of stay, and number of searches of the keywords of interest in the historical browsing behavior data.

10. A retrieval device, characterized in that, include: The first determination module is used to determine multiple target objects in the database that correspond to the object to be retrieved. The second determining module is used to determine the scene pattern features corresponding to each target object based on keyword data; the keyword data includes comment data and attribute data. The third determining module is used to determine the user's attention scene characteristics based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data. The comparison module is used to compare the features of the scene of interest with the features of the scene pattern to obtain the similarity between each feature of the scene pattern and the features of the scene of interest. The fourth determining module is used to determine the target display method based on the similarity, and to display the target object in the search results according to the target display method.

11. A terminal, characterized in that, include: processor; The processor is used to determine multiple target objects in the database that correspond to the object to be retrieved; The scene pattern features corresponding to each target object are determined based on keyword data; the keyword data includes comment data and attribute data. The user's attention scenario characteristics are determined based on behavioral keyword data; the behavioral keyword data includes the user's browsing behavior data. The similarity between the scene features of interest and the scene pattern features is obtained by comparing each scene pattern feature with the scene features of interest. The target display method is determined based on the similarity, and the target object is displayed in the search results according to the target display method.

12. A terminal, characterized in that, include: Transceiver, processor, memory, and programs or instructions stored in the memory and executable on the processor; When the processor executes the program or instructions, it implements the steps of the method as described in any one of claims 1 to 9.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 9.

14. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 9.

Images