Information recommendation method and device, and electronic device

Abstract

The present disclosure provides an information recommendation method and device and electronic equipment, relates to the technical field of data processing, and in particular to the technical field of search recommendation. The specific implementation scheme is: obtaining target search information input by a user, the target search information comprising M words, M being a positive integer; determining a first category corresponding to the target search information and a first core word in the M words; obtaining historical search information associated with the first category in a first database to obtain first candidate recommendation information, the first database being used to store categories and historical search information in association; obtaining a descriptive word associated with the first core word in a second database; and splicing the first core word and the descriptive word associated with the first core word to obtain second candidate recommendation information, the second database being used to store core words and descriptive words in association; and performing information recommendation based on the first candidate recommendation information and the second candidate recommendation information.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of data processing, in particular to the technical field of search recommendation, and more particularly to an information recommendation method and device and electronic equipment. BACKGROUND

[0002] With the development of science and technology and Internet technology, e-commerce platforms have rapidly emerged. E-commerce platforms can provide users with various network services, bringing great convenience to production and life.

[0003] For example, users can purchase goods on e-commerce platforms. With the search capabilities of e-commerce platforms, when users query information on e-commerce platforms, e-commerce platforms usually recommend information to users according to the search information input by users to provide more accurate demand expression for users.

[0004] Currently, e-commerce platforms usually recommend information based on a single category of goods or a single core word of search information. SUMMARY

[0005] The present disclosure provides an information recommendation method, device and electronic equipment.

[0006] According to a first aspect of the present disclosure, an information recommendation method is provided, comprising:

[0007] obtaining target search information input by a user, the target search information comprising M words, M being a positive integer;

[0008] determining a first category corresponding to the target search information and a first core word in the M words;

[0009] obtaining historical search information associated with the first category in a first database to obtain first candidate recommendation information, the first database being associated with storage of categories and historical search information;

[0010] obtaining a description word associated with the first core word in a second database; and splicing the first core word and the description word associated with the first core word to obtain second candidate recommendation information, the second database being associated with storage of core words and description words;

[0011] performing information recommendation based on the first candidate recommendation information and the second candidate recommendation information.

[0012] According to a second aspect of the present disclosure, an information recommendation device is provided, comprising:

[0013] a first obtaining module configured to obtain target search information input by a user, the target search information comprising M words, M being a positive integer;

[0014] The first determining module is configured to determine a first category corresponding to the target search information and a first core word in the M keywords;

[0015] The second obtaining module is configured to obtain historical search information associated with the first category in a first database to obtain first candidate recommended information, wherein the first database stores categories and historical search information in association.

[0016] The third obtaining module is configured to obtain a descriptive word associated with the first core word in a second database, and splice the first core word and the descriptive word associated with the first core word to obtain second candidate recommended information, wherein the second database stores core words and descriptive words in association.

[0017] The recommending module is configured to perform information recommendation based on the first candidate recommended information and the second candidate recommended information.

[0018] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0019] at least one processor; and

[0020] a memory connected with the at least one processor in communication; wherein

[0021] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the methods in the first aspect.

[0022] According to a fourth aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, the computer instructions being used to enable a computer to perform any one of the methods in the first aspect.

[0023] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements any one of the methods in the first aspect.

[0024] The technical solution of the present disclosure solves the problem of poor information recommendation effect and improves the effect of information recommendation.

[0025] It should be understood that the contents described in this part are not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0026] The accompanying drawings are used to better understand the present scheme and do not limit the present disclosure. Among them:

[0027] Figure 1is a flowchart of an information recommendation method according to a first embodiment of the present disclosure;

[0028] Figure 2 is a structural diagram of an information recommendation device according to a second embodiment of the present disclosure;

[0029] Figure 3 is a schematic block diagram of an example electronic device to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0030] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, which should be considered in their context only. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0031] First Embodiment

[0032] As shown in Figure 1 The present disclosure provides an information recommendation method, comprising the following steps:

[0033] Step S101: obtaining target search information input by a user, the target search information comprising M words.

[0034] Wherein, M is a positive integer.

[0035] In this embodiment, the information recommendation method relates to the field of data processing technology, and particularly relates to the field of search recommendation technology, and can be widely applied in a commodity procurement scenario. The information recommendation method of the present embodiment can be executed by the information recommendation device of the present embodiment. The information recommendation device of the present embodiment can be configured in any electronic device to execute the information recommendation method of the present embodiment. The electronic device can be a server or a terminal device, which is not limited here.

[0036] The present embodiment can be applied to a toB procurement scenario or a toC procurement scenario, which is not limited here. The difference between the toB procurement scenario and the toC procurement scenario lies in the different users that the e-commerce platform faces. For the toB procurement scenario, it usually faces groups such as communities, groups or units, while for the toC procurement scenario, it usually faces individuals.

[0037] In step S101, the target search information can be information input by a user in a search input box, such as the user inputting the target search information "notebook" on an e-commerce platform.

[0038] The target search information can include M words, where M is a positive integer. The M words can include nouns, adjectives, or verbs, etc. For example, if the target search information is "small notebook", the target search information can include an adjective and a noun. For another example, if the target search information is "folding airplane", the target search information can include a verb and a noun.

[0039] In an optional embodiment, if the user inputs the target search information by separating different words with spaces, the information recommendation apparatus can divide the words by detecting the spaces to obtain the M words.

[0040] In another optional embodiment, the information recommendation apparatus can perform word segmentation on the target search information by using a pre-trained word segmentation model, such as the jieba word segmentation tool, to obtain the M words and the word segmentation weight of each word. Before word segmentation, the words in a pre-stored stop word list can be used to filter out the useless characters in the target search information, and then the word segmentation is performed, which can improve the accuracy of the word segmentation.

[0041] There can be multiple occasions for obtaining the target search information. For example, the information recommendation apparatus can obtain the target search information in real time or periodically. For another example, when it is detected that the user is performing input in the search input box, the information recommendation apparatus can be triggered to obtain the target search information. For another example, when the user confirms that the information input in the search input box is complete, the information recommendation apparatus can be triggered to obtain the target search information. For another example, when the user enters a product detail page or finishes browsing the product detail page, the information recommendation apparatus can be triggered to obtain the target search information input by the user before, which can be the search information that can search for the product detail page.

[0042] In an optional embodiment, when the information recommendation apparatus detects that the user is performing input in the search input box, the information recommendation apparatus can obtain the target search information input by the user at present, and perform information recommendation based on the target search information. In this embodiment, the information recommendation apparatus can predict the search information that the user can need based on the target search information, and can display the predicted search information in the search input box in the form of a drop-down menu to perform information recommendation, so as to provide more accurate demand expression for the user.

[0043] In another optional embodiment, when the information recommendation apparatus detects that the user enters or browses a commodity detail page, the target search information input by the user previously can be acquired, and information recommendation is performed based on the target search information and the commodity detail page entered based on the target search information. In this embodiment, the information recommendation apparatus can predict the search information that the user can need, and can display the predicted search information in a preset position, such as a right position of the page, to perform information recommendation, so as to provide more accurate demand expression for the user.

[0044] Step S102: determining a first category corresponding to the target search information and a first core word in the M words.

[0045] In this step, the first category can refer to a commodity category corresponding to the target search information. For example, the target search information is "notebook", and the commodity category corresponding to the target search information can include stationery.

[0046] The first category can refer to a first-level category, a second-level category, or a third-level category of a commodity corresponding to the target search information. In order to accurately express the commodity category corresponding to the target search information and accurately search for commodities according to the category, the first category can specifically refer to a third-level category of a commodity corresponding to the target search information. For example, the target search information is "notebook", the commodity category corresponding to the target search information from the first-level category to the third-level category is stationery, notebook, and notebook respectively, the third-level category of the commodity corresponding to the target search information is notebook, and the first category corresponding to the target search information can be notebook.

[0047] The determination manner of the first category can include multiple manners. For example, the target search information can be subjected to category analysis to obtain first target category information of the target search information, and a category in the first target category information can be determined as the first category.

[0048] For another example, if the information recommendation apparatus detects that the user enters a first page (which can be a commodity detail page) based on the target search information, the first category can be determined based on category information obtained by category analysis of the target search information and a first category list corresponding to an object (which can be a commodity) associated with the first page. In an optional embodiment, a category in the intersection of the category information obtained by the category analysis of the target search information and the first category list can be determined as the first category.

[0049] For another example, if the target search information matches a target historical search information in a first database, a category associated with the target historical search information can be determined as the first category, wherein the first database stores categories and historical search information in association.

[0050] The first core word in the M words can refer to a keyword in the M words, which can be a noun and refers to a word in the target search information that best expresses the user's search demand. For example, if the target search information is "hard shell notebook", the keyword of the target search information is notebook.

[0051] The determination manner of the first core word can also include various manners. For example, it can be determined whether a word in the M words matches a word in a keyword library, and the word in the M words that matches the keyword library is determined as the first core word.

[0052] For example, for each word in the M words, the word can be analyzed by category to obtain category information corresponding to the word; at least one candidate word is determined from the M words, and the category information corresponding to each candidate word intersects with the category information obtained by analyzing the target search information by category; and the first core word is determined from the at least one candidate word.

[0053] Step S103: Obtain the historical search information associated with the first category in the first database to obtain the first candidate recommendation information, and the first database stores the category and the historical search information in association.

[0054] In this step, the first database stores the category and the historical search information in association, and one category can store one, two or more historical search information in association. The historical search information associated with the first category in the first database can be obtained to obtain the first candidate recommendation information.

[0055] Wherein, before the information recommendation, the category corresponding to the historical search information in the log data can be determined in advance, and the category and the historical search information are stored in association in the first database, which will be described in detail below.

[0056] Step S104: Obtain the description word associated with the first core word in the second database; and splice the first core word and the description word associated with the first core word to obtain the second candidate recommendation information, and the second database stores the core word and the description word in association.

[0057] In this step, the second database stores the core word and the description word in association, and one core word can store one, two or more description words in association. The description word associated with the first core word in the second database can be obtained, and the first core word and the description word associated with the first core word can be spliced to obtain the second candidate recommendation information.

[0058] For example, the first core word is a notebook, the description words associated with the first core word in the second database include hard shell, small size, super thickness, etc. The first core word is spliced with each description word to obtain the second candidate recommended information, including the spliced search information "hard shell notebook", "small size notebook" and "super thickness notebook".

[0059] In the information recommendation, the core word in the historical search information in the log data can be determined in advance, the words in the historical search information other than the core word are determined as the description words, and the core word and the description words are associated and stored in the second database, which will be described in detail below.

[0060] Step S105: information recommendation based on the first candidate recommended information and the second candidate recommended information.

[0061] In this step, the first candidate recommended information can include one, two or more recommended search information, and each recommended search information is the historical search information associated with the first category. The second candidate recommended information can also include one, two or more recommended search information, and each recommended search information is the information obtained by splicing the first core word and the description words associated with the first core word.

[0062] In an optional embodiment, the first candidate recommended information and the second candidate recommended information can be summarized. After summarization, there may be an intersection between the first candidate recommended information and the second candidate recommended information, i.e. there may be duplicate recommended search information. In this case, de-duplication can be performed, and only one of the duplicate recommended search information needs to be saved.

[0063] In this embodiment, the recommendation weight of each recommended search information can be determined, and the recommended search information can be sorted according to the recommendation weight from large to small. The recommended search information with a high recommendation weight is recommended to the user.

[0064] If the recommended search information is the recommended search information in the first candidate recommended information, the weight corresponding to the recommended search information can be determined, for example, the weight corresponding to the recommended search information can be determined as the recommendation weight of the recommended search information. In the first database, each historical search information associated with a category can correspond to a weight. The greater the weight, the closer the association between the historical search information and the category.

[0065] If the to-be-recommended search information is to-be-recommended search information in the second candidate recommended information, the weight corresponding to the descriptive word in the to-be-recommended search information can be determined, for example, the weight corresponding to the descriptive word in the to-be-recommended search information can be determined as the recommended weight of the to-be-recommended search information. In the second database, each descriptive word associated with a core word can correspond to a weight, and the greater the weight, the closer the association between the descriptive word and the core word.

[0066] If the to-be-recommended search information is to-be-recommended search information in the intersection of the first candidate recommended information and the second candidate recommended information, in this case, the weight corresponding to the to-be-recommended search information and the weight corresponding to the descriptive word in the to-be-recommended search information can be determined, for example, the average of the weight corresponding to the to-be-recommended search information and the weight corresponding to the descriptive word in the to-be-recommended search information can be determined as the recommended weight of the to-be-recommended search information.

[0067] In another optional implementation, the first recommended weight of the historical search information in the first candidate recommended information can be determined based on a preset first channel weight, the second recommended weight of the information obtained by splicing the first core word and the descriptive word associated with the first core word in the second candidate recommended information can be determined based on a preset second channel weight, the target recommended information can be determined from the first candidate recommended information and the second candidate recommended information based on the first recommended weight and the second recommended weight, and information recommendation can be performed based on the target recommended information.

[0068] In this implementation, the first channel weight can be a channel for information recommendation based on categories, the second channel weight can be a channel for information recommendation based on core words, and the first channel weight and the second channel weight can be preset, for example, the first channel weight can be set to 0.7, and the second channel weight can be set to 0.3.

[0069] Correspondingly, the first recommended weight can be determined based on the first channel weight and the weight corresponding to the historical search information, i.e., to-be-recommended search information, in the first candidate recommended information, and the second recommended weight can be determined based on the second channel weight and the weight corresponding to the descriptive word in the to-be-recommended search information in the second candidate recommended information. For example, the two weights can be multiplied to obtain the recommended weight of the to-be-recommended search information.

[0070] In this embodiment, by acquiring target search information input by a user, the target search information including M words; determining a first category corresponding to the target search information and a first core word in the M words; acquiring historical search information associated with the first category in a first database to obtain first candidate recommendation information, the first database being associated with storage of categories and historical search information; acquiring a descriptive word associated with the first core word in a second database; and splicing the first core word and the descriptive word associated with the first core word to obtain second candidate recommendation information, the second database being associated with storage of core words and descriptive words; and performing information recommendation based on the first candidate recommendation information and the second candidate recommendation information. In this way, the recommendation of search guidance can be performed in combination with the two channels of the category and the core word corresponding to the target search information, so that the effect of information recommendation can be improved.

[0071] It should be noted that the third candidate recommendation information can be obtained by combining other channels, such as acquiring historical search information matched with the target search information from log data, and the third candidate recommendation information can be combined with the first candidate recommendation information and the second candidate recommendation information for information recommendation.

[0072] Optionally, the determination of the first category corresponding to the target search information includes any one of the following:

[0073] The target search information is analyzed to obtain first target category information, the first target category information including the first category;

[0074] In a case where the target search information enters a first page, the first category is determined based on a first category list corresponding to an object associated with the second target category information and the first page, at least one of the second target category information and the first category list including the first category;

[0075] In a case where the target search information matches target historical search information in the first database, the category associated with the target historical search information is determined as the first category.

[0076] In this embodiment, in one scenario, if the target search information is search information input by a user for the first time, there is no historical search information matching the target search information in the first database and the log data. In this scenario, the information recommendation device can analyze the target search information, and specifically, a pre-trained category analysis model can be used to analyze the target search information to obtain first target category information of the target search information.

[0077] The category analysis model can use the third-level category as the analysis target, that is, the category in the first target category information obtained by analysis is a third-level category. The first target category information can include at least one category and the weight corresponding to each category. Correspondingly, the category with the largest weight in the first target category information can be determined as the first category.

[0078] In another scenario, if the user inputs the target search information and enters the first page based on the target search information, the first page can be a commodity detail page. In this case, the second target category information (which can be the category information obtained by analyzing the target search information, that is, the first target category information) can be combined with the first category list corresponding to the object associated with the first page to determine the first category. In this way, the commodity category that the user may need can be adjusted in real time based on the user's search situation, thereby providing more accurate demand expression for the user.

[0079] In this scenario, the first category list corresponding to the object associated with the first page can also include at least one category. The category in the first category list can also be a third-level category. The category of the object (which can be a commodity) associated with the commodity detail page can be determined by analyzing the content of the commodity detail page to obtain the first category list.

[0080] In a possible implementation, the category with the largest weight in the intersection of the second target category information and the first category list can be determined as the first category.

[0081] In another scenario, if the target historical search information matching the target search information can be found in the first database, the category associated with the target historical search information can be determined as the first category in this scenario.

[0082] In this embodiment, the first category corresponding to the target search information can be determined in various ways. The commodity category that the user may need can be adjusted in real time based on the user's search situation, thereby providing more accurate demand expression for the user and further improving the effect of information recommendation.

[0083] Optionally, M is greater than 1, and the first core word in the M words is determined based on the following manner:

[0084] For each word in the M words, the word is subjected to category analysis to obtain category information corresponding to the word;

[0085] At least one candidate word is determined from the M words, and the category information corresponding to each candidate word intersects with the category information obtained by analyzing the target search information;

[0086] The first core word is determined from the at least one candidate word.

[0087] In this embodiment, for each of the M words, the word can be analyzed by category to obtain category information corresponding to the word. The category analysis manner can be similar to the target search information category analysis manner, which is not limited here.

[0088] The category information corresponding to the word can include at least one category and a weight corresponding to each category. At least one candidate word can be determined from the M words. The determination manner can be that, for each of the M words, it can be determined whether the category information corresponding to the word has an intersection with the category information (which can be the first target category information) obtained by analyzing the target search information by category. If there is an intersection, the word can be determined as a candidate word.

[0089] Then, the first core word can be determined from the at least one candidate word. Specifically, the target weight of each candidate word can be determined. In an optional embodiment, the segmentation weight and the position weight of the candidate word can be multiplied to obtain the target weight of the candidate word. The position weight of the candidate word can be weighted according to the position of the candidate word in the target search information. The earlier the position, the greater the position weight. Accordingly, the candidate word with the maximum target weight in the at least one candidate word can be determined as the first core word.

[0090] In this embodiment, the first core word in the target search information is determined by combining the category, which can improve the accuracy of the core word determination, and thus can further improve the effect of information recommendation.

[0091] It should be noted that the determination manner of the first core word can also be used to determine the core word in the second database. Specifically, log data can be obtained, which can include historical search information. The historical search information in the log data can be filtered by using a pre-established search information blacklist list, and the historical search information in the log data can be filtered by using a pre-established category blacklist list. For example, the historical search information corresponding to the category information in the log data that has an intersection with the category blacklist list can be filtered out. Meanwhile, the useless characters of the historical search information can be filtered out by using the words in the pre-stored stop word list.

[0092] For the filtered log data, a segmentation model can be used to segment the historical search information in the log data to obtain segmentation results and segmentation weights. If the segmentation result is only one, the segmentation result is determined as the core word, if the segmentation result has multiple, the category information of each segmentation result and the category information of the historical search information are cross-validated to obtain candidate words. If the candidate words are only one, the candidate words are the core words, if the candidate words have multiple, the candidate words are weighted according to the position sequence in the historical search information, the target weight of the candidate words = segmentation weight * position ratio, the candidate words with the maximum target weight are taken as the core words, and the remaining segmentation results of the historical search information are taken as the description words.

[0093] After that, if the core words of different historical search information in the log data are the same, the description words associated with the core words can be clustered, and finally a core word can be associated with multiple description words, and the core word and the description word are stored in the second database. In addition, the weight of the description word can be determined according to the segmentation weight and the position ratio of the description word, and the weight can be stored in the second database.

[0094] Optionally, the step S105 specifically includes:

[0095] Based on the preset first channel weight, a first recommendation weight of the historical search information in the first candidate recommendation information is determined;

[0096] Based on the preset second channel weight, a second recommendation weight of the information obtained by splicing the first core word and the description words associated with the first core word in the second candidate recommendation information is determined;

[0097] Based on the first recommendation weight and the second recommendation weight, a target recommendation information is determined from the first candidate recommendation information and the second candidate recommendation information;

[0098] Based on the target recommendation information, information recommendation is performed.

[0099] In the embodiment, the first channel weight can be a channel for information recommendation based on category, and the second channel weight can be a channel for information recommendation based on core word. The first channel weight and the second channel weight can be preset, for example, the first channel weight can be set to 0.7, and the second channel weight can be set to 0.3.

[0100] The first recommendation weight can be determined based on the first channel weight and a weight corresponding to the history search information, i.e., the to-be-recommended search information, in the first candidate recommendation information, and the second recommendation weight can be determined based on the second channel weight and a weight corresponding to the description word in the to-be-recommended search information in the second candidate recommendation information. For example, the two weights can be multiplied to obtain the recommendation weight of the to-be-recommended search information.

[0101] In a possible implementation, the to-be-recommended search information in the first candidate recommendation information can be sorted in descending order of the first recommendation weight, and the to-be-recommended search information with a higher first recommendation weight is determined as the target recommendation information. The to-be-recommended search information in the second candidate recommendation information can also be sorted in descending order of the second recommendation weight, and the to-be-recommended search information with a higher second recommendation weight is determined as the target recommendation information.

[0102] When information recommendation is performed based on the determined target recommendation information, if there is repeated search information in the determined target recommendation information, a de-duplication process can be performed, and then the target recommendation information is recommended to the user.

[0103] In this embodiment, the search guidance recommendation can be performed in combination with the category and the core word corresponding to the target search information, so that the effect of information recommendation can be improved.

[0104] Optionally, before the step S103, the method further includes:

[0105] The history search information associated with the second page is obtained from the log data;

[0106] The history search information is parsed to obtain third target category information;

[0107] The category corresponding to the history search information is determined based on the third target category information and a second category list corresponding to the object associated with the second page;

[0108] The category corresponding to the history search information is stored in the first database in association with the history search information.

[0109] In this embodiment, the log data can be obtained, which can include the history search information. The history search information in the log data can be filtered by using a pre-established search information blacklist, and the history search information in the log data can be filtered by using a pre-established category blacklist, for example, the history search information corresponding to the category information intersecting with the category blacklist in the log data can be filtered out. Meanwhile, the useless characters of the history search information can be filtered out by using the words in the pre-stored stop word list.

[0110] For the filtered log data, historical search information associated with the second page in the log data can be obtained. The second page can refer to a commodity detail page, and when entering the commodity detail page based on the historical search information, the historical search information can be associated with the commodity detail page.

[0111] The historical search information is subject to category analysis to obtain third target category information. The manner of subjecting the historical search information to category analysis is similar to the manner of subjecting the target search information to category analysis, which will not be repeated here.

[0112] In the case where the third target category information includes at least two categories, the third target category information can be calibrated in combination with a second category list corresponding to an object associated with the second page to obtain a category corresponding to the historical search information, which can be a strong category expression of the historical search information. The manner of obtaining the second category list corresponding to the object associated with the second page is the same as that of obtaining the first category list corresponding to the object associated with the first page, which will not be repeated here.

[0113] The number of second pages is one, two or more, which will not be specifically limited. In the case where the number of second pages is more, the commodity category results of all second pages can be combined together to obtain the second category list.

[0114] In an optional embodiment, if the third target category information and the second category list corresponding to the object associated with the second page have an intersection, the category corresponding to the historical search information can be the category in the intersection. In another optional embodiment, if the third target category information and the second category list corresponding to the object associated with the second page have no intersection, the category corresponding to the historical search information can be the category in the third target category information or the category in the second category list.

[0115] In the specific implementation process, in order to effectively establish the mapping of the category to the historical search information, the historical search information in the log data can be clustered according to the category field obtained by analyzing the historical search information, so that the same category can correspond to multiple historical search information. Correspondingly, all historical search information with the same category corresponding to the historical search information in the log data can be clustered together, and the category and the historical search information are associated and stored in the first database.

[0116] In the first database, the historical search information associated with the category can be sorted according to the weight of the category field obtained by analyzing the historical search information, and the weight of the category field obtained by analyzing the historical search information can be stored in the first database as the weight corresponding to the historical search information.

[0117] In this embodiment, the third target category information obtained by analyzing the historical search information and the second category list of the second page associated with the historical search information are combined to determine the category corresponding to the historical search information, so as to determine the strong category expression of the historical search information, and the strong category expression of the historical search information is stored in the first database in association with the historical search information. In this way, the problem of inaccurate category caused by determining the category of search information only by relying on the commodity category can be solved, the accuracy of determining the category of search information can be improved, and the effect of information recommendation can be improved.

[0118] Optionally, the third target category information includes at least one category and a weight corresponding to each category, and the category corresponding to the historical search information is determined based on the third target category information and the second category list corresponding to the object associated with the second page, including at least one of the following:

[0119] In the case where the third target category information and the second category list do not have an intersection, the category with the highest frequency of occurrence in the second category list is determined as the category corresponding to the historical search information.

[0120] In the case where the third target category information and the second category list have an intersection, if the weight of the target category is less than a preset threshold, the category with the largest weight in the third target category information is determined as the category corresponding to the historical search information, and the target category is a category in the intersection of the third target category information and the second category list.

[0121] In the case where the third target category information and the second category list have an intersection, if the weight of the target category is greater than or equal to the preset threshold, the target category is determined as the category corresponding to the historical search information.

[0122] In this embodiment, the category corresponding to the historical search information is one, which can be the strongest category expression of the historical search information.

[0123] Specifically, in one scenario, if the third target category information and the second category list do not have an intersection, the number of occurrences of each category in the second category list can be calculated, and the category with the highest frequency of occurrence in the second category list is determined as the result of the final category calibration, and the category corresponding to the historical search information is obtained.

[0124] In another scenario, if the third target category information and the second category list have an intersection, the target category can be used as a candidate category, wherein the target category can be a category in the intersection of the third target category information and the second category list, and the number of target categories can be one, two or more.

[0125] Afterwards, it can be determined whether the weight of the target category is greater than or equal to a preset threshold value. If it is determined that the weight of all target categories is less than the preset threshold value, such as 0.5, it is determined that the commodity category result is divergent, and the category value is difficult to believe. The category with the largest weight in the third target category information is determined as the final category calibration result, and the category corresponding to the historical search information is obtained.

[0126] If there is a target category with a weight greater than the preset threshold value, such as 0.5, in the intersection, the target category can be the final category calibration result of the historical search information corresponding to the historical search information.

[0127] In this way, the third target category information obtained by analyzing the historical search information and the second category list of the second page associated with the historical search information can be combined to determine the strongest category expression of the historical search information, further improving the accuracy of the search information category determination.

[0128] Second embodiment

[0129] As shown in Figure 2 The present disclosure provides an information recommendation device 200, comprising:

[0130] A first acquisition module 201 is configured to acquire target search information input by a user, wherein the target search information comprises M words, and M is a positive integer;

[0131] A first determination module 202 is configured to determine a first category corresponding to the target search information and a first core word in the M words;

[0132] A second acquisition module 203 is configured to acquire historical search information associated with the first category in a first database to obtain first candidate recommendation information, wherein the first database stores categories and historical search information in association;

[0133] A third acquisition module 204 is configured to acquire a description word associated with the first core word in a second database, and splice the first core word and the description word associated with the first core word to obtain second candidate recommendation information, wherein the second database stores core words and description words in association;

[0134] A recommendation module 205 is configured to perform information recommendation based on the first candidate recommendation information and the second candidate recommendation information.

[0135] Optionally, the first determination module 202 comprises:

[0136] A first analysis unit is configured to analyze the target search information to obtain first target category information, wherein the first target category information comprises the first category;

[0137] The first determining unit is configured to, in a case where the target search information enters a first page, determine the first category based on a second target category information and a first category list corresponding to an object associated with the first page, at least one of the second target category information and the first category list including the first category;

[0138] The second determining unit is configured to, in a case where the target search information matches target historical search information in the first database, determine a category associated with the target historical search information as the first category.

[0139] Optionally, M is greater than 1, and the first determining module 202 includes:

[0140] The second parsing unit is configured to, for each word in the M words, perform category parsing on the word to obtain category information corresponding to the word.

[0141] The third determining unit is configured to determine at least one candidate word from the M words, each candidate word corresponding to category information intersecting with category information obtained by performing category parsing on the target search information.

[0142] The fourth determining unit is configured to determine the first core word from the at least one candidate word.

[0143] Optionally, the recommendation module 205 is specifically configured to:

[0144] determine a first recommendation weight of the historical search information in the first candidate recommendation information based on a preset first channel weight;

[0145] determine a second recommendation weight of information obtained by splicing the first core word and a descriptive word associated with the first core word in the second candidate recommendation information based on a preset second channel weight;

[0146] determine target recommendation information from the first candidate recommendation information and the second candidate recommendation information based on the first recommendation weight and the second recommendation weight;

[0147] recommend information based on the target recommendation information.

[0148] Optionally, the apparatus further includes:

[0149] The fourth obtaining module is configured to obtain historical search information from log data, the historical search information being associated with a second page.

[0150] The category parsing module is configured to perform category parsing on the historical search information to obtain third target category information.

[0151] The second determining module is configured to determine the category corresponding to the historical search information based on the third target category information and a second category list corresponding to the object associated with the second page.

[0152] The storage module is configured to store the category corresponding to the historical search information in association with the historical search information in the first database.

[0153] Optionally, the third target category information includes at least one category and a weight corresponding to each category, and the second determining module includes:

[0154] The fifth determining unit is configured to, in a case where the third target category information and the second category list do not have an intersection, determine a category with the most occurrences in the second category list as the category corresponding to the historical search information.

[0155] The sixth determining unit is configured to, in a case where the third target category information and the second category list have an intersection, determine a category with the largest weight in the third target category information as the category corresponding to the historical search information, if a weight of a target category is less than a preset threshold, the target category being a category in the intersection of the third target category information and the second category list.

[0156] The seventh determining unit is configured to, in a case where the third target category information and the second category list have an intersection, determine the target category as the category corresponding to the historical search information, if the weight of the target category is greater than or equal to the preset threshold.

[0157] The information recommendation apparatus 200 provided by the present disclosure can implement each process of the information recommendation method embodiment and achieve the same beneficial effects. To avoid repetition, details are not described herein.

[0158] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0159] According to the embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium and a computer program product.

[0160] Figure 3A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0161] As shown in Figure 3 The device 300 includes a computing unit 301 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 302 or a computer program loaded into a random access memory (RAM) 303 from a storage unit 308. In the RAM 303, various programs and data required for the operation of the device 300 can also be stored. The computing unit 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0162] Various components in the device 300 are connected to the I / O interface 305, including an input unit 306, such as a keyboard, a mouse, and the like; an output unit 307, such as various types of displays, speakers, and the like; the storage unit 308, such as a magnetic disk, a magneto-optical disk, and the like; and a communication unit 309, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 309 allows the device 300 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0163] The computing unit 301 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 301 performs various methods and processes described above, such as the information recommendation method. For example, in some embodiments, the information recommendation method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 300 via the ROM 302 and / or the communication unit 309. When the computer program is loaded onto the RAM 303 and executed by the computing unit 301, one or more steps of the information recommendation method described above can be performed. Alternatively, in other embodiments, the computing unit 301 can be configured to perform the information recommendation method by any other appropriate means, such as by means of firmware.

[0164] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0165] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine or entirely on a remote machine or server.

[0166] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0167] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; 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 acoustic, speech, or tactile input.

[0168] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0169] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0170] It should be understood that the various forms of flow shown above can be used to reorder, add, or delete steps. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure can be achieved, which is not limited herein.

[0171] The above detailed description does not constitute a limitation on the protection scope of the present disclosure. 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 replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. An information recommendation method, comprising: obtaining target search information input by a user, the target search information comprising M terms, M being a positive integer; determining a first category corresponding to the target search information and a first core term in the M terms; obtaining historical search information associated with the first category in a first database to obtain first candidate recommendation information, the first database being associated with storage of categories and historical search information; obtaining a description term associated with the first core term in a second database; and splicing the first core term and the description term associated with the first core term to obtain second candidate recommendation information, the second database being associated with storage of core terms and description terms; performing information recommendation based on the first candidate recommendation information and the second candidate recommendation information; before the obtaining of the historical search information associated with the first category in the first database, the method further comprises: obtaining historical search information from log data, the historical search information being associated with a second page; performing category analysis on the historical search information to obtain third target category information; determining a category corresponding to the historical search information based on the third target category information and a second category list corresponding to an object associated with the second page; storing the category corresponding to the historical search information and the historical search information in the first database in association; the third target category information comprises at least one category and a weight corresponding to each category, and the determination of the category corresponding to the historical search information based on the third target category information and the second category list corresponding to the object associated with the second page comprises at least one of the following: in a case where there is no intersection between the third target category information and the second category list, determining a category with the highest occurrence frequency in the second category list as the category corresponding to the historical search information; in a case where there is an intersection between the third target category information and the second category list, if a weight of a target category is less than a preset threshold, determining a category with the largest weight in the third target category information as the category corresponding to the historical search information, the target category being a category in the intersection between the third target category information and the second category list; in a case where there is an intersection between the third target category information and the second category list, if the weight of the target category is greater than or equal to the preset threshold, determining the target category as the category corresponding to the historical search information.

2. The method of claim 1, wherein, the determination of the first category corresponding to the target search information comprises any of the following: performing category analysis on the target search information to obtain first target category information, the first target category information comprising the first category; in a case where the target search information enters a first page, determining the first category based on second target category information and a first category list corresponding to an object associated with the first page, at least one of the second target category information and the first category list comprising the first category; in a case where the target search information matches target historical search information in the first database, determining a category associated with the target historical search information as the first category.

3. The method of claim 1, wherein, M is greater than 1, a first core word in the M words is determined based on the following manner: For each word in the M words, the word is category parsed to obtain category information corresponding to the word; At least one candidate word is determined from the M words, and the category information corresponding to each candidate word intersects with the category information obtained by category parsing the target search information; The first core word is determined from the at least one candidate word.

4. The method of claim 1, wherein, The information recommendation based on the first candidate recommendation information and the second candidate recommendation information comprises: Based on the preset first channel weight, determine the first recommendation weight of the historical search information in the first candidate recommendation information; Based on the preset second channel weight, determine the second recommendation weight of the information obtained by splicing the first core word and the descriptive word associated with the first core word in the second candidate recommendation information; Based on the first recommendation weight and the second recommendation weight, determine the target recommendation information from the first candidate recommendation information and the second candidate recommendation information; Based on the target recommendation information, information recommendation is performed.

5. An information recommendation device, comprising: A first acquisition module for acquiring target search information input by a user, the target search information comprising M words, M being a positive integer; A first determination module for determining a first category corresponding to the target search information and a first core word in the M words; A second acquisition module for acquiring historical search information associated with the first category in a first database to obtain first candidate recommendation information, the first database being associated with storage of categories and historical search information; A third acquisition module for acquiring descriptive words associated with the first core word in a second database, and splicing the first core word and the descriptive words associated with the first core word to obtain second candidate recommendation information, the second database being associated with storage of core words and descriptive words; A recommendation module for information recommendation based on the first candidate recommendation information and the second candidate recommendation information; A fourth acquisition module for acquiring historical search information from log data, the historical search information being associated with a second page; A category parsing module for category parsing the historical search information to obtain third target category information; A second determination module for determining a category corresponding to the historical search information based on the third target category information and a second category list corresponding to an object associated with the second page; A storage module for storing the category corresponding to the historical search information and the historical search information in the first database; The third target category information comprises at least one category and a weight corresponding to each category, and the second determination module comprises: A fifth determination unit for determining, in the case that there is no intersection between the third target category information and the second category list, a category appearing most frequently in the second category list as the category corresponding to the historical search information; The sixth determination unit is configured to, in a case where the third target category information intersects with the second category list, determine, if a weight of a target category is less than a preset threshold, a category with a maximum weight in the third target category information as a category corresponding to the historical search information, the target category being a category in the intersection of the third target category information and the second category list; The seventh determination unit is configured to, in a case where the third target category information intersects with the second category list, determine, if the weight of the target category is greater than or equal to the preset threshold, the target category as the category corresponding to the historical search information.

6. The apparatus of claim 5, wherein, The first determination module comprises: The first analysis unit is configured to perform category analysis on the target search information to obtain first target category information, the first target category information comprising the first category; The first determination unit is configured to, in a case where the target search information enters a first page, determine the first category based on a second target category information and a first category list corresponding to an object associated with the first page, at least one of the second target category information and the first category list comprising the first category; The second determination unit is configured to, in a case where the target search information matches a target historical search information in the first database, determine a category associated with the target historical search information as the first category.

7. The apparatus of claim 5, wherein, M is greater than 1, and the first determination module comprises: The second analysis unit is configured to, for each word in the M words, perform category analysis on the word to obtain category information corresponding to the word; The third determination unit is configured to determine at least one candidate word from the M words, each candidate word corresponding to category information that intersects with category information obtained by performing category analysis on the target search information; The fourth determination unit is configured to determine the first core word from the at least one candidate word.

8. The apparatus of claim 5, wherein, The recommendation module is specifically configured to: determine, based on a preset first channel weight, a first recommendation weight of historical search information in the first candidate recommendation information; determine, based on a preset second channel weight, a second recommendation weight of information obtained by splicing the first core word and a descriptive word associated with the first core word in the second candidate recommendation information; determine, based on the first recommendation weight and the second recommendation weight, target recommendation information from the first candidate recommendation information and the second candidate recommendation information; recommend information based on the target recommendation information.

9. An electronic device comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable 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 of any one of claims 1-4.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-4.

11. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-4.

Images