Search recommendation method and apparatus, intelligent device, electronic device, and storage medium

Abstract

The application discloses a search recommendation method and device, intelligent equipment, electronic equipment and a storage medium, which can be applied to artificial intelligence, cloud computing, big data, computer data, intelligent search, information flow and knowledge graph. The specific implementation scheme is as follows: a search task of a user is acquired, a search target object is determined according to the search task, the category to which the search target object belongs is determined, and an associated category having an associated relationship with the category is determined; recommendation information is generated and output according to the search target object and the associated category, wherein the recommendation information includes the search target object and objects belonging to the associated category; by combining the objects belonging to the associated category for recommendation, the problem that the search recommendation in the related art has great limitations is avoided, technical effects of improving the flexibility, diversity, richness and reliability of the search recommendation are achieved, and the search experience of the user is enhanced.

Description

Technical Field

[0001] This application relates to computer and data processing technologies, and more particularly to a search recommendation method, apparatus, smart device, electronic device, and storage medium, which can be applied to artificial intelligence, cloud computing, big data, computer data, intelligent search, information flow, and knowledge graphs. Background Technology

[0002] With the development of internet technology and the increase in information volume, how to improve the comprehensiveness of information recommendations to users has become an urgent problem to be solved.

[0003] In existing technologies, the common approach is to identify objects with a high degree of similarity to the search target, and then generate and output recommendation information based on the search target and the identified objects with high similarity. For example, if the search target is a mobile phone model XX, and the identified objects with high similarity could be mobile phones model YY that have an appearance similar to model XX, then recommendation information is generated and output based on mobile phones model XX and model YY.

[0004] However, generating recommendations based on similarity has relatively large limitations and cannot meet the broad needs of users. Summary of the Invention

[0005] This application provides a search recommendation method, apparatus, smart device, electronic device, and storage medium for improving the comprehensiveness of recommendations.

[0006] According to a first aspect of this application, a search recommendation method is provided, comprising:

[0007] Obtain the user's search task and determine the search target object based on the search task;

[0008] Determine the category to which the search target object belongs, and determine the associated categories that are related to the category to which it belongs;

[0009] Based on the search target object and the associated category, recommendation information is generated and output; wherein, the recommendation information includes: the search target object, and objects belonging to the associated category.

[0010] The search and recommendation method provided in this application embodiment achieves the technical effect of improving the diversity and flexibility of recommendations.

[0011] According to a second aspect of this application, a search recommendation apparatus is provided, comprising:

[0012] The first acquisition module is used to acquire the user's search task and determine the search target object based on the search task;

[0013] The first determining module is used to determine the category to which the search target object belongs, and to determine the associated categories that are related to the category to which it belongs;

[0014] A generation module is used to generate recommendation information based on the search target object and the associated category; wherein, the recommendation information includes: the search target object, and objects belonging to the associated category;

[0015] The output module is used to output the recommendation information.

[0016] According to a third aspect of this application, an electronic device is provided, comprising:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in the first aspect.

[0020] According to a fourth aspect of this application, a smart device is provided, comprising:

[0021] The device includes an output device, at least one processor, and a memory communicatively connected to the at least one processor; wherein the output device is connected to the at least one processor.

[0022] The memory stores instructions that can be executed by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method as described in the first aspect;

[0023] The output device is used to output the recommendation information.

[0024] According to a fifth aspect of this application, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform the method as described in the first aspect.

[0025] According to a sixth aspect of this application, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0026] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0027] The accompanying drawings are provided for a better understanding of this solution and do not constitute a limitation of this application. Wherein:

[0028] Figure 1 This is a schematic diagram illustrating an application scenario according to an embodiment of this application;

[0029] Figure 2 This is a schematic diagram according to one embodiment of the present application;

[0030] Figure 3 This is a schematic diagram according to yet another embodiment of this application;

[0031] Figure 4 This is a schematic diagram of a visualization chart according to an embodiment of this application;

[0032] Figure 5 This is a schematic diagram of a visualization chart according to another embodiment of this application;

[0033] Figure 6 This is a schematic diagram of a visualization chart according to another embodiment of this application;

[0034] Figure 7 This is a schematic diagram according to yet another embodiment of this application;

[0035] Figure 8 This is a schematic diagram according to yet another embodiment of this application;

[0036] Figure 9 This is a schematic diagram according to yet another embodiment of this application;

[0037] Figure 10 This is a schematic diagram according to yet another embodiment of this application;

[0038] Figure 11 This is a schematic diagram according to yet another embodiment of this application. Detailed Implementation

[0039] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0040] In one example, the search recommendation method provided in this application embodiment can be applied to scenarios involving item recommendations (such as product recommendations). For instance, the server recommends furniture, appliances, and books to users.

[0041] In another example, the search recommendation method provided in this application embodiment can be applied to information recommendation (such as news recommendation) scenarios. For example, the server recommends the latest news, weather information, and traffic information to users.

[0042] In another example, the search recommendation method provided in this application embodiment can be applied to human-computer interaction scenarios. For instance, based on user-initiated interaction information, the server recommends topics and related content relevant to the interaction information to the interactive robot.

[0043] It is worth noting that the above examples are only for illustrative purposes, illustrating the applicable scenarios of the search recommendation method in this application embodiment, and should not be construed as limiting the application scenarios of the search recommendation method in this application embodiment.

[0044] To deepen the reader's understanding of the application scenarios of the search recommendation method in the embodiments of this application, we will now take item recommendation (such as product recommendation) as an example, and combine it with Figure 1 ( Figure 1 The following is a detailed description of the application scenarios of the search recommendation method according to the embodiments of this application (illustrated in the form of a schematic diagram of the application scenarios).

[0045] In such Figure 1 In the application scenario shown, when user 101 has a search need, such as the need to purchase goods, they can send a search request to server 103 based on user terminal 102.

[0046] For example, such as Figure 1 As shown, user 101 can enter "XX air conditioner" on the display interface of user terminal 102 and send a search request to server 103 through the virtual "Search" button on the display interface of user terminal 102.

[0047] Server 103 receives a search request sent by user terminal 102 for searching "XX air conditioner", generates and sends recommendation information including "XX air conditioner" back to user terminal 102.

[0048] User terminal 102 can display recommended information including "XX air conditioner".

[0049] In related technologies, in order to improve the user's shopping experience, server 103 can identify "YY Air Conditioner" which has a high similarity to "XX Air Conditioner", generate and feed back recommendation information including "XX Air Conditioner" and "YY Air Conditioner" to user terminal 102.

[0050] Among them, high similarity can characterize both high performance similarity and high appearance similarity.

[0051] In other words, in related technologies, while recommending products that users search for, servers can also recommend products related to those products.

[0052] However, recommending products to users based on similarity may have limitations and fail to meet the broad needs of users. In other words, using related technologies to recommend relevant information may result in low flexibility and comprehensiveness in the recommendations.

[0053] The inventors of this application, through creative work, arrived at the inventive concept of this application: determining the category to which the object to be recommended (such as an item) belongs, determining the related categories that are associated with the category to which it belongs, and generating recommendation information based on the object to be recommended and the related categories, thereby achieving the technical effect of flexible and comprehensive recommendation.

[0054] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0055] This application provides a search recommendation method, apparatus, smart device, electronic device, and storage medium, which are applied in the fields of artificial intelligence, cloud computing, big data, computer vision, intelligent search, and knowledge graphs in the technology of computer and data processing, so as to achieve the technical effects of flexible and comprehensive search recommendation.

[0056] Figure 2 This is a schematic diagram according to one embodiment of the present application, such as... Figure 2 As shown, the search recommendation method in this application embodiment may include:

[0057] S201: Obtain the user's search task and determine the search target based on the search task.

[0058] For example, the execution entity in this embodiment can be a search and recommendation device, and the search and recommendation device can be a server (including cloud servers and local servers), a computer, a terminal device, a processor, and a chip, etc.

[0059] For example, when the search recommendation method of this embodiment is applied to, such as Figure 1 In the application scenarios shown, the search recommendation device can be as follows: Figure 1 The server shown.

[0060] In one example, a search task can be initiated by the user. For instance, when a user has a search need, a search task is initiated to the search recommendation device based on the user's device (such as a mobile terminal).

[0061] In this embodiment, the search target can be the object that the user is searching for.

[0062] For example, combined with Figure 1 In the application scenario shown, when a user initiates a search task for "XX air conditioner" on the search recommendation device based on the user's device, the search recommendation device will identify "XX air conditioner" as the search target.

[0063] In another example, the search task can be generated by a search recommendation device based on the user's historical search records. For instance, the search recommendation device generates search tasks based on time intervals and the user's historical search records.

[0064] In this embodiment, the search target can be an object from the user's historical search history.

[0065] For example, if a user searches for "XX air conditioner" recently (e.g., within a week) but has not placed an order, the search recommendation device can proactively generate a search task and generate and send recommendation information to the user's device.

[0066] Furthermore, based on the search and recommendation device actively generating search tasks, the search and recommendation device can generate and send recommendation information to the user device when the relevant information of the search target object changes, such as price information.

[0067] It should be understood that the above examples are only for illustrative purposes and are not intended to be construed as limiting the triggering of search tasks or search tasks in general.

[0068] S202: Determine the category to which the target object belongs, and determine the related categories that are associated with the target category.

[0069] It is worth noting that the categories in this embodiment can be categories determined based on industry classification standards, such as home decoration as one category and automobiles as another; or they can be categories set based on needs, historical records, and experience, such as doors and windows as one category and home appliances as another.

[0070] In this embodiment, when the category is based on industry classification standards, the search recommendation method of this embodiment can be made relatively flexible and comprehensive.

[0071] Combining the above examples and Figure 1 In the application scenario shown, the search target is "XX air conditioner", which indicates that the category of the search target is home decoration.

[0072] Accordingly, after determining the primary category, we can then identify related categories that are associated with it. In other words, there is a relationship between the primary category and related categories, such as home improvement and automobiles.

[0073] S203: Generate and output recommendation information based on the search target and related categories.

[0074] The recommended information includes: the target search object, and objects belonging to related categories.

[0075] In this embodiment, the recommendation information may include two dimensions: one dimension is the target search object, and the other dimension is the objects belonging to the associated categories.

[0076] Based on the above example, if the search target is "XX air conditioner" and the associated category is automobile, the recommended information will include "XX air conditioner" and objects belonging to automobiles (such as "XX automobile").

[0077] For example, the search recommendation device can output recommendation information through an interactive interface. For instance, combined with... Figure 1 In the application scenario shown, the search and recommendation device can send recommendation information to the user terminal and output the recommendation information through the user terminal's interactive interface (i.e., display interface).

[0078] It is worth noting that in this embodiment, a feature is introduced to determine the category to which the search target object belongs and to identify the associated categories that are related to the target object. By introducing this feature, recommendation information that includes the search target object and objects belonging to associated categories can be generated. This allows the recommendation information to include more recommendation elements, thereby achieving the technical effects of comprehensiveness, flexibility, and reliability of the recommendation.

[0079] Figure 3 This is a schematic diagram according to yet another embodiment of this application, as shown below. Figure 3 As shown, the search recommendation method in this application embodiment may include:

[0080] S301: For each category of text data, perform object extraction processing on the text data of each category, and construct visualization charts including each category based on the objects obtained from the object extraction processing.

[0081] For example, text data can be obtained based on categories, and the text data for each category may include: the content of the text, the title of the text, comments about the object in the text, and descriptions about the object in the text, etc.

[0082] Object extraction processing can be achieved based on technologies such as named entity recognition. For example, nouns, place names, organization names, and product label names can be used as object terms for objects, such as car brand names in the context of automobiles, names of major tourist attractions, names of various departments, and so on.

[0083] In this embodiment, different objects can constitute a visual chart, and each category in the visual chart can include multiple different objects. That is, the visual chart includes multiple categories, and each category includes multiple different objects.

[0084] In some embodiments, there is a membership relationship between the objects of a visualization chart and the categories of visualization charts, that is, the objects of a visualization chart belong to the categories of visualization charts, and the membership relationship can be represented based on the second-side connection relationship.

[0085] Figure 4 This is a schematic diagram of a visualization chart according to an embodiment of this application, such as... Figure 4 As shown, the visualization chart can include two categories: home decoration and automobiles. "XX Air Conditioner" and "Frameless Wooden Doors and Windows" are objects of the visualization chart and both belong to the home decoration category; "Car A" and "Car B" are also objects of the visualization chart and both belong to the automobile category.

[0086] Figure 5 This is a schematic diagram of a visualization chart according to another embodiment of this application, such as... Figure 5 As shown, in Figure 4 Based on the visualization charts shown, in some embodiments, subcategories can be determined for each category. For example, subcategories under the home improvement category include: doors and windows and home appliances; subcategories under the automobile category include: gasoline vehicles and new energy vehicles.

[0087] And such as Figure 5 As shown, the subcategory "Doors and Windows" can include "Frameless Wooden Doors and Windows" objects and "Framed Wooden Doors and Windows" objects; the subcategory "Home Appliances" can include "XX Air Conditioner" objects and "XX Refrigerator" objects; the subcategory "Fuel Vehicles" can include "C Cars" and "D Cars"; and the subcategory "New Energy Vehicles" can include "A Cars" and "B Cars".

[0088] Similarly, subcategories can be determined based on industry standards, or based on demand, historical records, and experiments.

[0089] S302: Based on the obtained historical search records, determine the co-occurrence frequency of each category.

[0090] Co-occurrence frequency represents the frequency of different categories appearing in adjacent search records in historical search records.

[0091] For example, historical search records can be preset within a time period, such as historical search records within six months.

[0092] Frequency can represent the number of times different categories appear repeatedly in adjacent search records in historical search records.

[0093] For example, if the object "XX Air Conditioner" belongs to the home decoration category and "A Car" belongs to the car category, and in the historical search records of the past six months, "XX Air Conditioner" and "A Car" are adjacent search records n times (n is an integer greater than or equal to 1), then the co-occurrence frequency of the home decoration category and the car category is n times.

[0094] S303: Determine the first edge connection relationship for each category based on the co-occurrence frequency, and update the visualization chart based on the first edge connection relationship for each category.

[0095] The first edge connection represents the association relationship.

[0096] For example, this embodiment can be understood as determining whether there is a relationship between two categories based on co-occurrence frequency, and when there is a relationship between two categories, it can be represented in a visualization chart based on the first edge connection relationship.

[0097] Based on the above examples, if a correlation is determined between the home decoration category and the automobile category based on co-occurrence frequency, then as follows: Figure 6 ( Figure 6 As shown in the schematic diagram of a visualization chart according to another embodiment of this application, it can be used as follows: Figure 5 Based on the above, add a first-edge connection between the home improvement category and the automotive category in the visualization chart.

[0098] It should be understood that the above examples are merely illustrative using two categories as examples, and should not be construed as limiting the number of categories.

[0099] It is worth noting that, in this embodiment, by determining the association between categories based on co-occurrence frequency, the accuracy and reliability of the determined association between categories can be improved. This can then be achieved by increasing the content dimension of the recommendation information when determining recommendation information based on a visual chart with association, thereby improving the flexibility, comprehensiveness, and reliability of the recommendation and enhancing the user experience.

[0100] S304: Obtain the user's search task and determine the search target based on the search task.

[0101] For example, the description of S304 can be found in S201, and will not be repeated here.

[0102] S305: Based on the visualization chart, determine the category to which the target object belongs if it has a second-sided connection with the search target object.

[0103] As can be seen from the above examples, there is a second-sided connection between the object and the category of the visualization chart. The second-sided connection represents the relationship between the object and the category. In this embodiment, when the search recommendation device determines the search target object based on the search task, it can determine the category that has a second-sided connection with the search target object from the visualization chart based on the search target object.

[0104] For example, combined with Figure 6 The visualization shown indicates that if the search target is "XX air conditioner," then "XX air conditioner" can be a home improvement category. Furthermore, as... Figure 6 As shown, the search recommendation device can determine the home electronics category to which "XX air conditioner" belongs, and then determine the home decoration category to which it belongs based on the home electronics category.

[0105] It is worth noting that, in this embodiment, by determining the category to which the target object belongs through a second-sided connection from the visualization chart, the efficiency and reliability of determining the category to which the target object belongs can be improved.

[0106] S306: Based on the visualization chart, identify the related categories that have a first-edge connection with their respective categories.

[0107] As can be seen from the above examples, there is a first-edge connection relationship between the categories of the visualization chart. The first-edge connection relationship represents the association relationship between the categories. In this embodiment, when the search recommendation device determines the category to which the search target object belongs based on the visualization chart, it can determine the associated category that has a first-edge connection relationship with the belonging category from the visualization chart based on the belonging category. That is, it can determine the associated category that has an association relationship with the belonging category from the visualization chart.

[0108] For example, combined with Figure 6 The visualization chart shown shows that if the category is home decoration, the search recommendation device can determine that the associated category is automobile based on the first-side connection between the home decoration category and the automobile category.

[0109] Similarly, by identifying associated categories that have a first-edge connection with their respective categories from visual charts, the efficiency and reliability of identifying associated categories can be improved. Furthermore, when combining associated categories to generate recommendation information, the flexibility and comprehensiveness of recommendations for users based on that information can be enhanced.

[0110] S307: Based on the visualization chart, identify objects that have a second-sided connection relationship with the associated category.

[0111] Combining the above examples and Figure 6 If the associated category is automobile, then the objects that have a second-side connection relationship with automobile can be one or more of "A automobile", "B automobile", "C automobile", and "D automobile".

[0112] S308: Generate and output recommendation information based on objects that have a second-sided connection with the associated category and the search target object.

[0113] The recommended information includes: the target search object, and objects belonging to related categories.

[0114] Combining the above examples and Figure 6 If the objects with a second-side connection relationship with the associated category are "Car A", "Car B", "Car C" and "Car D", and the search target object is "Air Conditioner XX", then the recommended information may include "Air Conditioner XX" and "Car A", or "Air Conditioner XX" and "Car B", or "Air Conditioner XX", "Car A", "Car B", "Car C", and so on.

[0115] It is worth noting that in this embodiment, by identifying objects that have a second-sided connection with the associated category, and generating and outputting recommendation information based on these objects and the search target object, the content dimension of the recommendation information can be increased, thereby meeting the user's broader search needs. Furthermore, by making recommendations for users based on associated categories, the recommendation information can be highly aligned with the user's search needs, thereby improving the technical effect of reliability and accuracy of the recommendation.

[0116] Figure 7 This is a schematic diagram according to yet another embodiment of this application, as shown below. Figure 7 As shown, the search recommendation method in this application embodiment may include:

[0117] S701: Obtain historical search records, determine the co-occurrence frequency of each category based on the historical search records, and construct a mapping relationship based on the co-occurrence frequency.

[0118] Co-occurrence frequency represents the frequency of different categories appearing in adjacent search records in historical search history. Mapping relationship represents the association between categories.

[0119] For example, the description of co-occurrence frequency can be found in the example above, and will not be repeated here.

[0120] In this embodiment, a mapping relationship can be constructed based on co-occurrence frequency to characterize the association between categories. In some embodiments, the mapping relationship can be represented by a linked list.

[0121] It is worth noting that in this embodiment, by determining the co-occurrence frequency based on historical search records and constructing a mapping relationship based on the co-occurrence frequency, the mapping relationship can more accurately express the association between categories, thereby achieving the technical effect of improving the accuracy and reliability of recommendation information when determining recommendation information based on the mapping relationship.

[0122] In some embodiments, constructing a mapping relationship based on co-occurrence frequency may include the following steps:

[0123] Step 1: Determine the probability of association between each category based on the co-occurrence frequency.

[0124] Among them, the correlation probability is the probability that, in historical search records, after the first category is searched, the second category will continue to be searched.

[0125] In other words, in this embodiment, the categories with a relationship are directional.

[0126] Combining the above examples and Figure 6 If the co-occurrence frequency of the home decoration category and the car category is n times, and the number of times the car category is searched after searching the home decoration category in the historical search records is m times, then the probability of association between the home decoration category and the car category = m / n, and correspondingly, the probability of association between the car category and the home decoration category = (nm) / n.

[0127] Step 2: Construct a mapping relationship based on the correlation probability.

[0128] After determining the relevance probability, the search recommendation device can construct a mapping relationship based on the relevance probability. For example, a relevance probability threshold can be preset; if the determined relevance probability between categories is greater than the relevance probability threshold, then a mapping relationship between the categories is constructed.

[0129] Specifically, in conjunction with the above example, if m / n is greater than the correlation probability threshold, a mapping relationship is constructed that includes the correlation between the home decoration category and the car category. In the mapping relationship, the home decoration category and the car category are directional, that is, when the category to which the home decoration belongs is determined, the car category can be determined to be associated with the home decoration category.

[0130] For example, the correlation probability threshold can be set based on demand, historical records, and experiments.

[0131] It is worth noting that, in this embodiment, by determining the association between categories based on the correlation probability, the reliability and accuracy of the determined association can be improved, making the association highly consistent with the user's search needs. This achieves the technical effect of improving the flexibility and reliability of the recommendation by ensuring that the objects recommended to the user are those that meet the user's search needs.

[0132] S702: Obtain the user's search task and determine the search target based on the search task.

[0133] For example, the description of S702 can be found in S101, and will not be repeated here.

[0134] S703: Based on the mapping relationship, determine the associated categories that are related to the category to which the category belongs.

[0135] It is worth noting that there are relationships between different categories, and the technical effect of improving the efficiency of determining related categories can be achieved by determining the related categories based on the mapping relationship.

[0136] S704: Obtain secondary search records based on the search target object.

[0137] The secondary search record represents the record of continuing to search for objects after searching for the target object.

[0138] Similarly, secondary search records are search records within a preset time period, such as search records within one month. The specific settings can be based on needs, historical records, and experiments, and this embodiment does not impose any limitations.

[0139] S705: Based on the associated category, select objects belonging to the associated category from the secondary search records.

[0140] For example, if the search target is "XX air conditioner", which belongs to the home decoration category, and the related category of the home decoration category is automobile, and the secondary search records include "XX mattress", "XX tile", and "A car", then the search recommendation device can identify "A car" as an object belonging to the related category.

[0141] In some embodiments, the search recommendation device may be based on, for example, Figure 6 The visualization shown selects objects belonging to the associated category from secondary search records.

[0142] It is worth noting that, in this embodiment, by selecting objects belonging to the associated category from the secondary search records, the accuracy and reliability of the recommendation information can be improved when generating recommendation information based on objects belonging to the associated category, thus making it highly likely that the recommendation information will closely match the user's search needs.

[0143] In some embodiments, S705 may include the following steps:

[0144] Step 1: From the secondary search records, select the object with the most search volume that belongs to the related category.

[0145] Step 2: Identify the objects with the most searches as belonging to the related categories.

[0146] Based on the above example, if the associated category is automobile, and the secondary search records include "Automobile A", "Automobile B", and "Automobile C" belonging to the automobile category, then the search recommendation device can determine the search counts of "Automobile A", "Automobile B", and "Automobile C". For example, if the search count of "Automobile A" is 'a', the search count of "Automobile B" is 'b', and the search count of "Automobile C" is 'c', and it is determined that the largest of a, b, and c is 'a', then the search recommendation device can select "Automobile A" as the target.

[0147] Accordingly, when the search recommendation device determines that the most frequently searched item belonging to the associated category is 'a', then "Car A" can be identified as an object belonging to the associated category.

[0148] It is worth noting that, in this embodiment, by determining the object with the most searches belonging to the associated category as the object belonging to the associated category, the high correlation between the objects belonging to the associated category and the search target object can be improved, making the generated recommendation information highly relevant to the user's search needs, thereby improving the technical effect of improving the accuracy and reliability of the recommendation.

[0149] It is worth noting that the method for determining the objects belonging to the related category can also be the objects with the most purchases; the objects with the most transactions; the objects with the best evaluation; and so on. This embodiment does not limit the method.

[0150] S706: Generate and output recommendation information based on the search target object and objects belonging to related categories.

[0151] Based on the above example, search recommendations can generate output recommendation information based on the search target object "XX air conditioner" and the object belonging to the related category "A car".

[0152] It is worth noting that in related technologies, the recommendation information output by the search recommendation device only includes the search target object "XX air conditioner" or other air conditioners with similar functions (or similar appearance) to the search target object "XX air conditioner". However, in this embodiment, the recommendation information also includes the object "A car" belonging to the associated category, thereby avoiding the limitations of the recommendations in related technologies and improving the technical effect of recommendation flexibility and comprehensiveness.

[0153] Figure 8 This is a schematic diagram according to yet another embodiment of this application, as shown below. Figure 8 As shown, the search recommendation device 800 in this application embodiment may include:

[0154] The first acquisition module 801 is used to acquire the user's search task and determine the search target object based on the search task.

[0155] The first determining module 802 is used to determine the category to which the search target object belongs, and to determine the associated category that is related to the category to which it belongs.

[0156] The generation module 803 is used to generate recommendation information based on the search target object and the associated category; wherein the recommendation information includes: the search target object and objects belonging to the associated category.

[0157] Output module 804 is used to output the recommendation information.

[0158] In some embodiments, there are associations between different categories; the first determining module 802 is used to determine the associated categories that are associated with the category to which the category belongs, based on a pre-set mapping relationship, wherein the mapping relationship represents the association between categories.

[0159] Figure 9 This is a schematic diagram according to yet another embodiment of this application, as shown below. Figure 9 As shown, in Figure 8 Based on the embodiments shown, the search recommendation device of this application embodiment may include:

[0160] The second acquisition module 805 is used to acquire historical search records.

[0161] The second determining module 806 is used to determine the co-occurrence frequency of each category based on the historical search records; wherein the co-occurrence frequency represents the frequency of different categories appearing in adjacent search records in the historical search records.

[0162] The first construction module 807 is used to construct the mapping relationship based on the co-occurrence frequency.

[0163] In some embodiments, the first construction module 806 is configured to determine the correlation probability between the categories based on the co-occurrence frequency, and construct the mapping relationship based on the correlation probability; wherein the correlation probability represents the probability that, after the first category is searched in the historical search records, the second category is searched.

[0164] In some embodiments, different categories constitute a visualization chart, and different categories in the visualization chart have a first edge connection relationship, which represents an association relationship; the first determining module 802 is used to determine, according to the visualization chart, the associated category that has a first edge connection relationship with the category to which it belongs.

[0165] Figure 10 This is a schematic diagram according to yet another embodiment of this application, as shown below. Figure 10 As shown, in Figure 9 Based on the embodiments shown, the search recommendation device of this application embodiment may include:

[0166] Extraction module 808 is used to perform object extraction processing on text data of each category.

[0167] The second construction module 809 is used to construct various types of visual charts based on the objects obtained from object extraction processing;

[0168] The third determining module 810 is used to determine the co-occurrence frequency of each category based on the acquired historical search records, and to determine the first edge connection relationship of each category according to the co-occurrence frequency; wherein, the co-occurrence frequency represents the frequency of different categories appearing in adjacent search records in the historical search records;

[0169] Update module 811 is used to update the visualization chart based on the first connection relationship of each category.

[0170] In some embodiments, different objects constitute a visualization chart, and the categories in the visualization chart have multiple different objects. There is a second-side connection relationship between the objects in the visualization chart and the categories in the visualization chart. The second-side connection relationship represents the belonging relationship between the object and the category. The first determining module 802 is used to determine the belonging category that has a second-side connection relationship with the search target object based on the visualization chart.

[0171] In some embodiments, the generation module 803 is configured to determine, based on the visualization chart, objects that have a second-side connection relationship with the associated category, and generate the recommendation information based on the objects that have a second-side connection relationship with the associated category and the search target object.

[0172] In some embodiments, the generation module 803 is used to obtain secondary search records based on the search target object; wherein, the secondary search records represent records of continuing to search for objects after searching the search target object, and based on the association category, selecting objects belonging to the association category from the secondary search records, and generating the recommendation information based on the search target object and the objects belonging to the association category.

[0173] In some embodiments, the generation module 803 is used to select the object with the most search times belonging to the associated category from the secondary search records, and determine the object with the most search times as the object belonging to the associated category.

[0174] According to embodiments of this application, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect above, for example, implementing... Figure 2 , Figure 3 as well as Figure 7 Training method of the facial expression prediction model shown in any of the embodiments

[0175] According to embodiments of this application, this application also provides an electronic device and a readable storage medium.

[0176] Electronic devices are intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0177] Figure 11 This is a schematic diagram according to yet another embodiment of this application, as shown below. Figure 11 As shown, the electronic device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access memory (RAM) 1103. The RAM 1103 may also store various programs and data required for the operation of the device 1100. The computing unit 1101, ROM 1102, and RAM 1103 are interconnected via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0178] Multiple components in device 1100 are connected to I / O interface 1105, including: input unit 1106, such as keyboard, mouse, etc.; output unit 1107, such as various types of monitors, speakers, etc.; storage unit 1108, such as disk, optical disk, etc.; and communication unit 1109, such as network card, modem, wireless transceiver, etc. Communication unit 1109 allows device 1100 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0179] The computing unit 1101 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the various methods and processes described above, such as the search recommendation method. For example, in some embodiments, the search recommendation method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1100 via ROM 1102 and / or communication unit 1109. When the computer program is loaded into RAM 1103 and executed by the computing unit 1101, one or more steps of the search recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform the search recommendation method by any other suitable means (e.g., by means of firmware).

[0180] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0181] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0182] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0183] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0184] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (LANs), blockchain-based service networks (BSNs), wide area networks (WANs), and the Internet.

[0185] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and Virtual Private Servers (VPS) services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0186] According to another aspect of the embodiments of this application, an intelligent device is also provided, comprising:

[0187] The device includes an output device, at least one processor, and a memory communicatively connected to the at least one processor; wherein the output device is connected to the at least one processor.

[0188] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to perform the method described in any of the above embodiments;

[0189] The output device is used to output the recommendation information.

[0190] In some embodiments, the smart device further includes: a receiver;

[0191] The receiver is used to receive a search request, which is used to request a search for the target object.

[0192] Alternatively, the receiver may be used to receive a click request, which is used to request the application to be opened.

[0193] In some embodiments, the receiver is any one of the following: a microphone, a smart screen, and a user interface.

[0194] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0195] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A search recommendation method, comprising: Obtain historical search records, and determine the co-occurrence frequency of the first industry category and the second industry category based on the historical search records; The co-occurrence frequency characterization refers to the frequency of different industry categories appearing in adjacent search records in the historical search records. The correlation probability between the first industry category and the second industry category is determined based on the co-occurrence frequency. The correlation probability represents the probability that, after the first industry category is searched in the historical search records, the second industry category will continue to be searched. If the correlation probability is greater than the correlation probability threshold, then it is determined that the first industry category and the second industry category are correlated. Obtain the user's search task and determine the search target object based on the search task; The first industry category to which the search target object belongs is determined, and a second industry category that is associated with the first industry category is determined based on the association relationship; wherein, the industry category is divided based on industry classification standards; each industry category includes multiple subcategories, and each subcategory includes different objects; Obtain secondary search records based on the search target object. The secondary search records represent records of continuing to search for objects after searching for the search target object within a preset time period. Based on the second industry category, select objects belonging to the second industry category from the secondary search records; Generate and output recommendation information, which includes: the search target object belonging to the first industry category and the object belonging to the second industry category.

2. The method according to claim 1, wherein determining the correlation between the first industry category and the second industry category includes: Construct a mapping relationship between the first industry category and the second industry category; Based on the aforementioned association, a second industry category is determined that is associated with the industry category to which the search target object belongs, including: Based on the mapping relationship, a second industry category that is associated with the industry category to which the search target object belongs is determined.

3. The method according to claim 1, further comprising: For the text data of each industry category, object extraction processing is performed on the text data of each industry category, and a visualization chart including each industry category is constructed based on the objects obtained from the object extraction processing. The determination of the correlation between the first industry category and the second industry category includes: Establish a first-edge connection between the first industry category and the second industry category in the visualization chart to obtain an updated visualization chart; Based on the aforementioned association, a second industry category is determined that is associated with the industry category to which the search target object belongs, including: Based on the updated visualization chart, a second industry category is determined that has a first-edge connection relationship with the industry category to which the search target object belongs.

4. The method according to claim 3, wherein the industry category in the visualization chart has multiple different objects, and there is a second-sided connection relationship between the objects in the visualization chart and the industry category in the visualization chart, wherein the second-sided connection relationship represents the affiliation relationship between the objects and the industry category.

5. The method according to claim 1, wherein, Based on the second industry category, objects belonging to the second industry category are selected from the secondary search records, including: From the secondary search records, select the object with the most searched times belonging to the related industry category, and determine the object with the most searched times as belonging to the related industry category.

6. A search recommendation device, comprising: The second acquisition module is used to acquire historical search records; The second determining module is used to determine the co-occurrence frequency of the first industry category and the second industry category based on the historical search records; The co-occurrence frequency characterization refers to the frequency of different industry categories appearing in adjacent search records in the historical search records. The first construction module is used to determine the correlation probability between the first industry category and the second industry category based on the co-occurrence frequency. The correlation probability represents the probability that, after the first industry category is searched in the historical search records, the second industry category will continue to be searched. If the correlation probability is greater than the correlation probability threshold, then it is determined that the first industry category and the second industry category are correlated. The first acquisition module is used to acquire the user's search task and determine the search target object based on the search task; The first determining module is used to determine the first industry category to which the search target object belongs, and to determine the second industry category that is associated with the first industry category based on the association relationship; wherein, the industry category is divided based on industry classification standards; each industry category includes multiple subcategories, and each subcategory includes different objects; The generation module is used to obtain secondary search records based on the search target object. The secondary search records represent records of continuing to search for objects after searching for the search target object within a preset time period. Based on the second industry category, select objects belonging to the second industry category from the secondary search records; Generate recommendation information; the recommendation information includes: the search target object belonging to the first industry category, and the object belonging to the second industry category.

7. The apparatus according to claim 6, The first construction module is also used to construct a mapping relationship between the first industry category and the second industry category; The first determining module is specifically used to determine a second industry category that is associated with the industry category to which the search target object belongs, based on the mapping relationship.

8. The apparatus according to claim 6, further comprising: The extraction module is used to perform object extraction processing on the text data of each industry category. The second building module is used to construct visualization charts for various industry categories based on the objects obtained from object extraction and processing; The update module is used to establish a first-edge connection relationship between the first industry category and the second industry category in the visualization chart to obtain an updated visualization chart; The first determining module is specifically used to determine, based on the updated visualization chart, a second industry category that has a first-edge connection relationship with the industry category to which the search target object belongs.

9. The apparatus according to claim 8, wherein, The industry category in the visualization chart contains multiple different objects. There is a second-sided connection relationship between the objects in the visualization chart and the industry category in the visualization chart. The second-sided connection relationship represents the relationship between the object and the industry category.

10. The apparatus according to claim 6, wherein, The generation module is used to select the object with the most search times belonging to the related industry category from the secondary search records, and determine the object with the most search times as the object belonging to the related industry category.

11. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.

12. A smart device, comprising: The device includes an output device, at least one processor, and a memory communicatively connected to the at least one processor; wherein the output device is connected to the at least one processor. The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to perform the method of any one of claims 1-5; The output device is used to output the recommendation information.

13. The intelligent device according to claim 12, further comprising: Receiver; The receiver is used to receive a search request, which is used to request a search for the target object. Alternatively, the receiver may be used to receive a click request, which is used to request the application to be opened.

14. The smart device according to claim 13, wherein the receiver is any one of the following: a microphone, a smart screen, and a user interface.

15. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-5.

16. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1-5.

Images