Object recommendation method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention discloses an object recommendation method and device, a storage medium and electronic equipment. Receiving an input first keyword through the input box, and determining a candidate search word matched with the first keyword; determining historical interaction information corresponding to the candidate search terms on the at least one user interaction behavior; on the basis of transaction information of a first candidate object similar to each candidate search word, transaction feature information of the candidate search words is determined; sorting the candidate search terms based on the transaction feature information and the historical interaction information corresponding to the candidate search terms; displaying the candidate search terms; in response to the keyword determination event, obtaining a target keyword corresponding to the keyword determination event, and determining a plurality of target objects matched with the target keyword; and recommending the target user based on the plurality of target objects. Therefore, the object recommendation efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to an object recommendation method, device, storage medium and electronic device. Background Art

[0002] With the rapid development of life and technology, people often use game applications for entertainment. In some UGC (User-generated Content) games, users need to search for required components and other objects by entering keywords.

[0003] In the process of research and practice of existing technologies, it is found that the existing object recommendation method only matches similar objects based on the keywords input by the user and recommends them to the user, and cannot accurately and efficiently recommend objects of interest to the user, resulting in low efficiency of object recommendation. Summary of the invention

[0004] The embodiments of the present application provide an object recommendation method, device, storage medium and electronic device, which can enable users to quickly and accurately determine the target keywords to be searched, and efficiently and accurately recommend objects of interest to users, thereby improving the efficiency of object recommendation.

[0005] The present application provides an object recommendation method, including:

[0006] receiving a first keyword inputted through an input box, and determining a candidate search term matching the first keyword;

[0007] Determining historical interaction information corresponding to the candidate search term in at least one user interaction behavior;

[0008] Determining transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each of the candidate search terms;

[0009] sorting the candidate search terms based on the transaction feature information and the historical interaction information corresponding to each of the candidate search terms;

[0010] displaying the candidate search terms;

[0011] In response to a keyword determination event, obtaining a target keyword corresponding to the keyword determination event, and determining a plurality of target objects matching the target keyword;

[0012] Recommendations are made based on the multiple target objects.

[0013] Accordingly, an embodiment of the present application provides an object recommendation device, including:

[0014] A receiving unit, configured to receive a first keyword inputted through an input box, and determine a candidate search term matching the first keyword;

[0015] A first determining unit, configured to determine historical interaction information corresponding to the candidate search term in at least one user interaction behavior;

[0016] A second determining unit, configured to determine transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each of the candidate search terms;

[0017] A sorting unit, configured to sort the candidate search terms based on the transaction feature information and the historical interaction information corresponding to each of the candidate search terms;

[0018] A display unit, used for displaying the candidate search terms;

[0019] A third determination unit, configured to, in response to a keyword determination event, obtain a target keyword corresponding to the keyword determination event, and determine a plurality of target objects matching the target keyword;

[0020] A recommendation unit is used to make recommendations based on the multiple target objects.

[0021] In addition, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and the computer program is suitable for loading by a processor to execute the steps in any object recommendation method provided in the embodiment of the present application.

[0022] In addition, an embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to implement the object recommendation method provided in the embodiment of the present application.

[0023] An embodiment of the present application also provides a computer program product, which includes a computer program, and the computer program is stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device performs the steps in the object recommendation method provided in the embodiment of the present application.

[0024] The embodiment of the present application receives an input first keyword through an input box, determines candidate search terms that match the first keyword; determines historical interaction information corresponding to the candidate search terms in at least one user interaction behavior; determines transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each candidate search term; sorts the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; displays the candidate search terms; in response to a keyword determination event, obtains a target keyword corresponding to the keyword determination event, and determines multiple target objects that match the target keyword; and makes recommendations based on the multiple target objects. In this way, by determining the candidate search terms associated with the first keyword currently entered by the user in the input box, and then sorting the candidate search terms according to the historical interaction information and transaction feature information corresponding to the candidate search terms, the candidate search terms are recommended to the user for selection based on the sorting results. This allows the user to quickly and accurately determine the target keyword to be searched, thereby determining multiple target objects matching the target keyword for recommendation to the user, and achieving efficient and accurate recommendation of objects of interest to the user. At the same time, recommending search terms based on transaction feature information can increase the probability of users purchasing the recommended target objects, thereby increasing the benefits brought in the search scenario, and further improving the efficiency of object recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 This is a schematic diagram of an implementation scenario of an object recommendation method provided in an embodiment of the present application;

[0027] Figure 2 It is a flowchart of an object recommendation method provided in an embodiment of the present application;

[0028] Figure 3 is a schematic diagram of a search term expansion of an object recommendation method provided in an embodiment of the present application;

[0029] Figure 4a It is a schematic diagram of a specific architecture of an object recommendation method provided in an embodiment of the present application;

[0030] Figure 4b is another specific architecture diagram of an object recommendation method provided in an embodiment of the present application;

[0031] Figure 4cIt is a schematic diagram of a search system architecture of an object recommendation method provided in an embodiment of the present application;

[0032] Figure 5 is a schematic diagram of the structure of an object recommendation device provided in an embodiment of the present application;

[0033] Figure 6 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0035] The present application provides an object recommendation method, device, storage medium and electronic device. The object recommendation device can be integrated in an electronic device, which can be a server or a terminal.

[0036] Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, network acceleration services (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms. Terminals may include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, etc. Terminals and servers can be directly or indirectly connected via wired or wireless communications, and this application does not limit this.

[0037] See also Figure 1 , taking the object recommendation device integrated into an electronic device as an example, Figure 1A schematic diagram of an implementation scenario of the object recommendation method provided in an embodiment of the present application, wherein the electronic device can be a server, and the electronic device can receive an input first keyword through an input box, determine a candidate search term that matches the first keyword; determine historical interaction information corresponding to the candidate search term in at least one user interaction behavior; determine transaction feature information of the candidate search term based on transaction information of a first candidate object similar to each candidate search term; sort the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; display the candidate search terms; in response to a keyword determination event, obtain a target keyword corresponding to the keyword determination event, and determine multiple target objects that match the target keyword; and make recommendations based on the multiple target objects.

[0038] It should be noted that Figure 1 The implementation environment scenario diagram of the object recommendation method shown is only an example. The implementation environment scenario of the object recommendation method described in the embodiment of the present application is to more clearly illustrate the technical solution of the embodiment of the present application, and does not constitute a limitation on the technical solution provided by the embodiment of the present application. It is known to those skilled in the art that with the evolution of data processing and the emergence of new business scenarios, the technical solution provided by the present application is also applicable to similar technical problems.

[0039] The solutions provided by the embodiments of the present application are specifically described by the following embodiments. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0040] This embodiment will be described from the perspective of an object recommendation device, which may be integrated into an electronic device, which may be a terminal and / or a server, and this application does not limit this.

[0041] See also Figure 2 , Figure 2 : is a flow chart of an object recommendation method provided in an embodiment of the present application. The object recommendation method includes:

[0042] In step 101, a first keyword inputted is received through an input box, and candidate search terms matching the first keyword are determined.

[0043] The first keyword may be a keyword currently input by the target user in the input box. The target user may be the user currently searching, that is, the target user may be the user who inputs the first keyword through the input box. The input box may be an area in the user terminal interface for inputting keywords for searching. For example, when the target user inputs "city" in the input box, the first keyword may be "city". For another example, when the target user inputs "happy" in the input box, the first keyword may be "happy". The candidate search term may be a search term similar to the first keyword. A plurality of search terms that may be searched by a user may be pre-set, so that a candidate search term matching the first keyword may be found in the search terms. The candidate search term may be a search term similar to the first keyword, or may be a search term containing the first keyword.

[0044] There are many ways to determine the candidate search terms that match the first keyword. For example, the similarity between multiple pre-stored search terms and the first keyword can be calculated, and the search terms with higher similarity can be determined as candidate search terms that match the first keyword.

[0045] For example, suppose that a search engine corresponding to a game has pre-stored multiple search terms, including search terms such as "red house", "happy kids", "happy Rubik's Cube", and "happy castle". When the target user enters the first keyword "happy", candidate search terms similar to the first keyword can be found in the pre-stored search terms, for example, candidate search terms such as "happy kids", "happy Rubik's Cube", and "happy castle".

[0046] In one embodiment, personalized search term recommendations can be made to the user based on the user's preferences before the user searches. For example, the second candidate object of interest to the target user and the target user's preference score for the second candidate object can be determined based on the target user's user association information; associated search terms associated with the second candidate object can be obtained; interest scores of the associated search terms can be determined based on the preference scores of the second candidate objects associated with each associated search term; associated search terms can be sorted based on their interest scores; and search term recommendations can be made through the target user's terminal based on the sorting results of the associated search terms.

[0047] Among them, the user association information can be the user's association information, for example, it can include information such as objects that the target user has clicked to view, objects that have been searched, objects that have been purchased, and the like. The object can be a virtual resource, for example, it can be a component, prop, virtual map, and the like. The component can include content that the user has customized and created. For example, in a UGC (User-Generated Content) game, the component can include objects such as a virtual character model, a castle, a mineral sapling, and a mineral tracker. Optionally, the user association information can be a user-related log of the target user. The second candidate object can be an object that the target user may be interested in based on the user association information. For example, when it is identified based on the user association information that the target user often clicks on components of multiple castle types, the second candidate object can be a castle-type component. When it is identified based on the user association information that the target user has purchased a mineral sapling component, the second candidate object can be a mineral sapling component, etc. The preference score may be information indicating the target user's preference for the second candidate object. For example, when it is identified based on the user association information that the target user clicked component 1 3 times and component 2 10 times, it may be indicated that the target user has a higher preference for component 2, and thus component 2 may be calculated to have a higher preference score. The associated search term may be a search term corresponding to the second candidate object. For example, assuming that the name of the second candidate object is "Happy Castle", the associated search term corresponding to the second candidate object may be "Happy Castle". For another example, the corresponding associated search term may be pre-set for the second candidate object. The interest score may be information indicating the target user's interest in the associated search term.

[0048] Optionally, a collaborative filtering algorithm can be used to calculate the target user's preference score for multiple preset search terms based on the target user's user association information, and then sort each search term according to the preference score, so as to determine the associated search terms that the target user is more likely to be interested in. By pushing the associated search terms to the target user's terminal for search term recommendation, personalized search term recommendations can be made to the user. The user obtains personalized recommended associated search terms before searching, which can enable the user to quickly and accurately obtain the search terms they need to search, thereby improving search efficiency.

[0049] In step 102, historical interaction information corresponding to the candidate search term in at least one user interaction behavior is determined.

[0050] Among them, the user interaction behavior can be the user's interaction behavior, for example, it can include click, add to shopping cart, favorite, purchase and other behaviors, and the historical interaction information can be information indicating the user's interaction behavior on the object recommended based on the candidate search term, for example, it can include click-through rate, add to cart rate, purchase rate and other information.

[0051] Among them, the click-through rate can be the probability that the user clicks on the object recommended based on the candidate search term after the candidate search term is searched. Specifically, the click-through rate can be the ratio of the number of clicks to the number of searches. The number of clicks can be the number of times the object recommended based on the candidate search term is clicked after the candidate search term is searched. The number of searches can be the number of times the candidate search term is searched as a search term.

[0052] The add-to-cart rate may be the probability that a user will add an object recommended based on the candidate search term to a shopping cart after the candidate search term is searched. Specifically, the add-to-cart rate may be the ratio of the number of add-to-cart times to the number of searches. The number of add-to-cart times may be the number of times an object recommended based on the candidate search term is added to a shopping cart after the candidate search term is searched. The number of searches may be the number of times the candidate search term is searched as a search term.

[0053] The purchase rate may be the probability that a user purchases an object recommended based on the candidate search term after the candidate search term is searched. Specifically, the purchase rate may be the ratio of the number of purchases to the number of searches. The number of purchases may be the number of times the object recommended based on the candidate search term is purchased after the candidate search term is searched. The number of searches may be the number of times the candidate search term is searched as a search term.

[0054] In step 103, transaction feature information of the candidate search terms is determined based on transaction information of the first candidate objects similar to each candidate search term.

[0055] The first candidate object may be an object similar to the candidate search term. For example, when the candidate search term is "castle", the first candidate object may be an object of various virtual castle types, such as a retro castle component, a pink castle component, a castle map, and the like. The transaction information may be information indicating the transaction attributes of the first candidate object. For example, the transaction information may include at least one of a transaction price, a buyer's location, a transaction time, and a transaction frequency. The transaction price may be the price of the first candidate object, the buyer's location may be the location of the user who purchased the first candidate object, the transaction time may be the time when the user purchased the first candidate object, and the transaction frequency may be the number of times the first candidate object was purchased within a preset period of time. The transaction feature information may be information indicating the transaction features of the first candidate object. The transaction features may include features such as transaction price, transaction concentration area, transaction concentration period, and transaction frequency. The transaction concentration area may be an area where buyers who purchased the first candidate object are concentrated. For example, for the first candidate object of the warm type, the buyers who may purchase it are mostly located in the north, and for the first candidate object of the moisture-proof type, the buyers who may purchase it are mostly located in the south. The transaction concentration period may be a period in which the first candidate objects are purchased. For example, the first candidate objects of the sunlight castle type may be purchased in the morning, and the first candidate objects of the darkness castle type may be purchased in the evening.

[0056] There are many ways to determine the transaction feature information of the candidate search terms based on the transaction information of the first candidate objects similar to each candidate search term. For example, the price feature information of the candidate search terms can be calculated based on the transaction price of the first candidate objects similar to each candidate search term; and / or, based on the buyer location of the first candidate objects similar to each candidate search term, the transaction concentration area of ​​the first candidate objects under the same candidate search term is determined, and the transaction area feature information of the candidate search term is obtained based on the transaction concentration area; and / or, based on the transaction time of the first candidate objects similar to each candidate search term, the transaction concentration time period of the first candidate objects under the same candidate search term is determined, and the transaction time period feature information of the candidate search term is obtained based on the transaction concentration time period; and / or, based on the transaction frequency of the first candidate objects similar to each candidate search term, the transaction frequency feature information of the candidate search term is calculated.

[0057] The price feature information may be information indicating the transaction price of the first candidate object, the transaction area feature information may be information indicating the transaction concentration area of ​​the first candidate object, the transaction period feature information may be information indicating the transaction concentration period of the first candidate object, and the transaction frequency feature information may be information indicating the transaction frequency of the first candidate object. Thus, based on the transaction feature information of the first candidate object, the probability of the first candidate object being purchased in different scenarios can be determined, so that based on the scenario where the target user is located, the first candidate object with a greater possibility of purchase in the current scenario can be recommended to the target user, so that the probability of the target user purchasing the first candidate object can be increased, thereby increasing the conversion rate of the first candidate object and increasing the benefits brought by the search scenario.

[0058] For example, when the target user searches for an object in the morning, the first candidate object with concentrated transactions in the morning can be recommended to the target user; when the target user is located in area A, the first candidate object with concentrated transactions in area A can be recommended to the target user. In addition, the first candidate object with a higher transaction frequency or a higher transaction price can also be recommended to the target user to increase the probability of the target user purchasing the first candidate object, thereby increasing the benefits brought in the object search scenario and thereby improving the object search efficiency.

[0059] In step 104, the candidate search terms are ranked based on the transaction feature information and historical interaction information corresponding to each candidate search term.

[0060] Among them, there are many ways to sort the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term. For example, the interaction score of each candidate search term can be calculated based on the historical interaction information corresponding to each candidate search term; the candidate search terms can be sorted based on the interaction score, and the target candidate search terms can be screened out from the sorted candidate search terms; the target candidate search terms can be sorted based on the transaction feature information of each target candidate search term to obtain the sorted target candidate search terms.

[0061] The interaction score may be information measuring the degree to which each candidate search word is liked by the user, and the target candidate search word may be a candidate search word that the user is more interested in.

[0062] There are multiple ways to calculate the interaction score of each candidate search term based on the historical interaction information corresponding to each candidate search term. For example, each type of historical interaction information corresponding to each candidate search term can be accumulated to obtain the interaction score of each candidate search term. For example, assuming that the click-through rate, add-to-cart rate, and purchase rate of candidate search term a are 0.4, each type of historical interaction information is accumulated to obtain the interaction score of candidate search term a of 0.7. assuming that the click-through rate, add-to-cart rate, and purchase rate of candidate search term b are 0.3, 0.1, and 0.05, each type of historical interaction information is accumulated to obtain the interaction score of candidate search term b of 0.45.

[0063] Optionally, a weight may be set for each type of historical interaction information, so that each type of historical interaction information corresponding to each candidate search term may be weighted and summed according to the weight corresponding to each type of historical interaction information to obtain the interaction score of each candidate search term. For example, assuming that the weight corresponding to the click-through rate is 1, the weight corresponding to the add-to-cart rate is 1.5, and the weight corresponding to the purchase rate is 2, and assuming that the click-through rate, add-to-cart rate, and purchase rate corresponding to the candidate search term s are 0.4, 0.2, and 0.1, then based on the weights, each type of historical interaction information is weighted and summed to obtain the interaction score of the candidate search term a of 0.4×1+0.2×1.5+0.1×2=0.9, and assuming that the click-through rate, add-to-cart rate, and purchase rate corresponding to the candidate search term b are 0.3, 0.1, and 0.05, then based on the weights, each type of historical interaction information is weighted and summed to obtain the interaction score of the candidate search term b of 0.3×1+0.1×1.5+0.05×2=0.55.

[0064] After calculating the interaction score of each candidate search term based on the historical interaction information corresponding to each candidate search term, the candidate search terms can be sorted based on the interaction score, and the target candidate search terms can be screened out from the sorted candidate search terms. There are many ways to sort the candidate search terms based on the interaction score and screen out the target candidate search terms from the sorted candidate search terms. For example, the candidate search terms can be sorted from high to low according to the interaction score, so that the target candidate search terms can be intercepted from the sorted candidate search terms according to the required number of target candidate search terms.

[0065] Among them, based on the transaction feature information of each target candidate search word, the target candidate search words are sorted, and there are many ways to obtain the sorted target candidate search words. For example, when the transaction feature information is the transaction price, the target candidate search words can be sorted from high to low according to the transaction price to obtain the sorted target candidate search words. When the transaction feature information is the transaction concentration area, the target candidate search words can be sorted from near to far according to the distance between the transaction concentration area and the user's location to obtain the sorted target candidate search words. When the transaction feature information is the transaction concentration period, the target candidate search words can be sorted from near to far according to the time difference between the transaction concentration period and the user's search time to obtain the sorted target candidate search words. When the transaction feature information is the transaction frequency, the target candidate search words can be sorted from high to low according to the transaction frequency to obtain the sorted target candidate search words.

[0066] In this way, by sorting the candidate search terms according to the transaction feature information and historical interaction information corresponding to each candidate search term and displaying them to the user, the user can quickly and intuitively obtain the candidate search terms that the user may be interested in, thereby increasing the possibility of the user purchasing objects recommended based on the candidate search terms, improving the search conversion rate, and then increasing the revenue in the search scenario.

[0067] In step 105, candidate search terms are displayed.

[0068] For example, candidate search terms may be displayed through the target user's terminal for the target user to select.

[0069] Among them, in the terminal interface of the target user, the candidate search terms can be displayed in the order of their ranking.

[0070] Optionally, the candidate search terms may also include popular search terms.

[0071] Optionally, after the candidate search terms are sorted based on the transaction feature information and historical interaction information of each target candidate search term to obtain the sorted target candidate search terms, the sorted target candidate search terms can be displayed through the terminal of the target user.

[0072] For example, see Figure 3 , Figure 3This is a schematic diagram of a search term expansion of an object recommendation method provided in an embodiment of the present application. When a target user enters the first keyword "Hello" in the input box, candidate search terms such as "Hello City", "Hello! Gentleman" and so on corresponding to the first keyword can be sorted and sent to the target user's terminal, so that the candidate search terms "Hello City", "Hello! Gentleman" and so on associated with the first keyword "Hello" entered by the target user can be displayed in the search interface of the terminal. In this way, the target user can quickly obtain the required search terms based on the associated candidate search terms without completely entering the required keywords, thereby improving the object search efficiency.

[0073] In step 106, in response to the keyword determination event, a target keyword corresponding to the keyword determination event is obtained, and a plurality of target objects matching the target keyword are determined.

[0074] The keyword determination event may be an event in which the target user determines the keyword to be searched, for example, the target user inputs a keyword in an input box and triggers a search operation, or the target user selects a candidate search term from candidate search terms associated with the first keyword based on the first keyword input. The target keyword may be a search term triggered by the target user to search. The target object may be an object similar to the target keyword.

[0075] In one embodiment, initial objects of multiple matching types may be obtained according to multiple matching dimensions during the recall phase. Then, during the offline sorting phase, business indicator considerations may be added to sort the initial objects according to the estimated revenue corresponding to the initial objects, so as to increase the game revenue in the search scenario.

[0076] Specifically, in response to a keyword determination event, before obtaining a target keyword corresponding to the keyword determination event and determining multiple target objects matching the target keyword, an initial object matching multiple preset search terms can be obtained based on at least one matching dimension; based on the priority corresponding to the matching dimension, the initial objects matching the search terms in each matching dimension are sorted to obtain the sorted initial objects; a second purchase probability and price corresponding to the initial object are determined; second estimated revenue information of the initial object is calculated based on the second purchase probability and price; based on the second estimated revenue information, the sorted initial objects under each matching dimension are sorted to obtain candidate objects corresponding to each search term.

[0077] Among them, the step of determining multiple target objects matching the target keyword may include: screening out target search terms matching the target keyword from the search terms; and determining multiple target objects matching the target keyword based on candidate objects corresponding to the target search terms.

[0078] Among them, the matching dimension may include dimensions such as complete matching, partial matching and similar matching. Different matching dimensions correspond to different matching degrees. Each matching dimension may be preset with a corresponding priority. The higher the matching degree, the higher the priority. For example, the priority of complete matching is greater than the priority of partial matching, and the priority of partial matching is greater than the priority of similar matching. The initial object may be an object that matches each search term and is recalled based on each matching dimension. The second purchase probability may be the probability that the initial object is purchased, that is, the probability that the initial object is purchased by all users. The second estimated revenue information may be information indicating the revenue that the initial object may bring. The target search term may be a search term in the preset search terms that matches the target keyword determined by the target user.

[0079] Among them, based on at least one matching dimension, there can be multiple ways to obtain the initial objects that match the preset multiple search terms. For example, an offline recall algorithm can be used, using content recall based on semantic matching, and the semantics of the search terms can be extracted by segmenting and embedding the search terms, so as to match the search terms and the objects to be recalled. Specifically, for a complete match, for example, from a mineral to a mineral tracker, a simple string match can be used. If the name of the object to be recalled has a complete match or inclusion relationship with the search term, it can be recalled as the initial object. For a partial match, for example, from a mineral tracker to a mineral seedling, segmentation can be performed based on the pool of items to be searched, and the importance of the segmentation can be calculated. The lower the frequency of occurrence, the higher the importance. Both the search terms and the items to be searched are segmented. If there are identical segmentations, they are marked as matches. The importance of the segmentations is added to obtain a matching index, and the recall is sorted and truncated according to the matching index. For similar matching, such as matching from the Old Summer Palace to the Forbidden City, a language representation model (Bidirectional Encoder Representations from Transformers, BERT) can be used to calculate the name of the object and the vectorized expression of the search term, so that the inner product of the vectorized expression of the search term and the object name can be calculated, that is, the matching degree of the search term and the object. Based on the matching degree, truncated recall is performed to obtain the corresponding initial object.

[0080] Among them, based on the priority corresponding to the matching dimension, the initial objects matched by the search term in each matching dimension are sorted, and there can be multiple ways to obtain the sorted initial objects. For example, assuming that based on the matching dimension of complete matching, 100 initial objects corresponding to the search term c are matched, based on the matching dimension of partial matching, 500 initial objects corresponding to the search term are matched, and based on the matching dimension of similar matching, 1000 initial objects corresponding to the search term are matched. Then, according to the priority corresponding to the matching dimension, the initial objects of the search term c under each matching dimension can be sorted, wherein the initial objects matched based on the matching dimension of complete matching are ranked 1-100, the initial objects matched based on the matching dimension of partial matching are ranked 101-600, and the initial objects matched based on the matching dimension of similar matching are ranked 601-1600.

[0081] Optionally, the second purchase probability may include at least one of a probability of the initial object being purchased in the search scenario, a probability of the initial object being purchased in all scenarios, and a probability of a similar object corresponding to the initial object being purchased. The similar object may be an object similar to the initial object.

[0082] There are many ways to determine the second purchase probability and price corresponding to the initial object. For example, relevant logs of all users can be obtained, and the second purchase probability corresponding to each initial object and the price of each initial object can be calculated based on the relevant logs.

[0083] There are many ways to calculate the second estimated revenue information of the initial object based on the second purchase probability and price. For example, the product of the second purchase probability and price corresponding to each initial object can be calculated to obtain the second estimated revenue information of each initial object.

[0084] Among them, based on the second estimated revenue information, the sorted initial objects under each matching dimension are sorted, and there can be multiple ways to obtain candidate objects corresponding to each search term. For example, the sorted initial objects under each matching dimension are sorted from high to low according to the corresponding second estimated revenue information to obtain candidate objects corresponding to each search term. For example, assuming that after the initial objects of the search term c under each matching dimension are sorted according to the priority corresponding to each matching dimension, the initial objects matched based on the matching dimension of complete match are ranked 1-100, the initial objects matched based on the matching dimension of partial match are ranked 101-600, and the initial objects matched based on the matching dimension of similar match are ranked 601-1600. Then, according to the second estimated revenue information corresponding to each initial object under the matching dimension of complete match, the arrangement order of each initial object ranked 1-100 under the matching dimension of complete match can be adjusted, according to the second estimated revenue information corresponding to each initial object under the matching dimension of partial match, the arrangement order of each initial object ranked 101-600 under the matching dimension of partial match can be adjusted, and according to the second estimated revenue information corresponding to each initial object under the matching dimension of similar match, the arrangement order of each initial object ranked 601-1600 under the matching dimension of similar match can be adjusted.

[0085] In one embodiment, taking the object as a component in a UGC game as an example, please refer to Figure 4a , Figure 4a It is a specific architecture diagram of an object recommendation method provided by an embodiment of the present application. Search word recall content, component recall content, component features, and player features can be stored in the data layer corresponding to the game, wherein the search word recall content can be the recalled search word, the component recall content can be the recalled component, the component feature can be the object feature, and the player feature can be the user feature. In this way, based on the data stored in the data layer, the search word can be scored based on the collaborative filtering algorithm, and the recall content can be mixed based on the rules, so that the candidate search word can be recommended to the user, thereby improving the efficiency of search word determination. In addition, the search word association and sorting can be performed according to the keywords input by the user. Specifically, the search word can be sorted based on the click-through rate corresponding to the search word, or sorted based on the purchase rate after the jump, and the recall content can be mixed based on the rules to further improve the search word input and acquisition efficiency. In addition, the recalled components can also be sorted offline, for example, they can be sorted based on the degree of semantic matching, and sorted based on the estimated revenue corresponding to each component, thereby improving the game revenue of the search scene.

[0086] In step 107 , recommendations are made based on multiple target objects.

[0087] Among them, there can be many ways to make recommendations based on multiple target objects. For example, the first purchase probability of the target user for the target object can be determined; based on the first purchase probability, the first estimated profit information of the target object can be determined; the target objects can be sorted based on the estimated profit information, and the target users can be recommended based on the sorted target objects.

[0088] There are many ways to determine the first purchase probability of the target user for the target object. For example, the user characteristics of the target user and the object characteristics of the target object can be obtained, and the first purchase probability of the target user purchasing the target object can be predicted based on the user characteristics, target keywords and the object characteristics of the target object.

[0089] The user feature may be information indicating the target user, the user's past behaviors constitute user tags, and the user feature is obtained based on these tags. The object feature may be information indicating the target object.

[0090] Optionally, an artificial intelligence model can be used to predict the first purchase probability of the target user purchasing the target object based on user characteristics, target keywords, and object characteristics of the target object. The artificial intelligence model may include a machine learning model (such as LightGBM) or a deep learning model (such as Bert).

[0091] Optionally, if the signal transmission quality of the target user's terminal meets a preset abnormal condition, a second purchase probability of the target object being purchased may be obtained as the first purchase probability of the target user for the target object.

[0092] The preset abnormal condition may determine that the signal transmission quality of the target user's terminal is abnormal, for example, it may include that the data packet transmission rate of the target user's terminal is less than a preset rate, etc. The second purchase probability includes at least one of the probability that the target object is purchased in the search scenario, the probability that the target object is purchased in all scenarios, and the probability that a similar object corresponding to the target object is purchased.

[0093] In this way, when the signal transmission quality of the target user's terminal is poor, the second purchase probability of the target object being purchased can be obtained based on the result of offline sorting, and the first estimated profit information of the target object can be calculated based on the second purchase probability, so as to sort the target objects based on the estimated profit information, thereby avoiding long response time in the object search process and improving the object search rate.

[0094] In one embodiment, when the signal transmission quality of the target user's terminal is poor, object recommendations can be made directly based on the offline sorting results to avoid long response times during the object search process and improve the object search experience.

[0095] There are many ways to determine the first estimated profit information of the target object based on the first purchase probability. For example, the price of the target object can be obtained, and the product of the price of the target object and the first purchase probability can be calculated to obtain the first estimated profit information of the target object.

[0096] Then, the target objects may be sorted from high to low based on the estimated revenue information, so that the sorted target objects may be pushed to the terminal of the target user for recommendation.

[0097] For example, taking objects as components, please refer to Figure 4b , Figure 4b It is another specific architecture diagram of an object recommendation method provided by an embodiment of the present application. The data layer can store component recall offline sorting content, component features and player features, wherein the component recall offline sorting content can be the candidate objects corresponding to each search term obtained in the recall stage and the offline sorting stage. Therefore, after the target user determines the target keyword, it can respond to the request message (i.e., http request) from the terminal to the server, perform online sorting, and sort based on the degree of semantic matching and the LightGBM model. That is, the LightGBM model can be used to determine the first purchase probability of the target user for the target object based on player features, target keywords and component features, and the first estimated revenue information of the target object can be determined based on the first purchase probability and the component price, so that the target objects can be sorted from high to low based on the estimated revenue information, so as to give priority to recommending the target objects with higher preset revenue to the target users based on the sorted target objects, so as to increase the game revenue brought in the search scenario, thereby improving the object recommendation efficiency.

[0098] In one embodiment, taking an object as a component as an example, please refer to Figure 4c , Figure 4cThis is a schematic diagram of the search system architecture of an object recommendation method provided in an embodiment of the present application. The data layer of the search system can store search association logs, post-search click purchase logs, player-related logs, and component-related logs. Based on these logs, the purchase rate of the component, as well as the transaction feature information and historical interaction information of each search term can be determined. In this way, multiple search terms and components of multiple matching dimensions can be recalled based on recall algorithms such as hot recall, collaborative filtering recall, and matching recall, as well as player features of players and component features of components in advance based on relevant logs through feature extraction algorithms, so that offline sorting can be performed based on the above data using a sorting algorithm, and sorting can be performed based on statistics plus rules, as well as sorting based on the second purchase rate estimated by a machine learning model, so that the search term recommendation results, association term recommendation results, and component recommendation results can be stored in a remote dictionary service (Remote Dictionary Server, Redis for short) database, so that when the target user determines the target keyword, the offline sorting results can be sorted online based on an online sorting algorithm, specifically, sorting can be performed based on statistics plus rules, as well as sorting based on the first purchase rate estimated by a machine learning model, and then, based on an http request triggered by the terminal based on an http interface, the corresponding search term recommendation interface, search input association interface, and component search interface can be called to push corresponding content to the user terminal, thereby realizing search term recommendation, search term association, and component recommendation.

[0099] In this way, the embodiment of the present application provides a fusion recommendation algorithm that uses search word recommendations, introduces input associations, integrates personalized recommendations, and covers the entire search process. First, it makes full use of the interface before and during the search, introduces search word recommendations and search input associations, and shortens the user's search path. Secondly, personalized search word recommendations are used based on user historical search big data to improve the user's search experience. In the sorting stage, the comprehensive benefits and semantic relevance of the search scene are considered, and the benefit indicators of the search scene and the search accuracy are improved. Through the optimization practice of the entire link of the search scene in the embodiment of the present application, more than 50% of players can be helped to improve their search experience through search word recommendations and search input associations, and the search intention can be achieved without entering all the search words. The number of characters entered per person is relatively reduced by 8.1%. The introduction of big data personalized recommendations helps players better find the products they intend. After iteration, the number of searches by players has decreased compared to before, and the average number of searches per purchase has been relatively reduced by 6.4%. In the sorting stage, semantic matching and product benefit indicators are considered in multiple dimensions. After strict online AB testing, the benefits of the search scene after optimization have increased by 5%.

[0100] In some sandbox games based on open UGC platforms, creators of the entire game world continue to produce high-quality content and fresh gameplay in the game, and build an open ecosystem with the majority of players and developers. The large number of components created by developers is an important part of the game, and the component search consumption flow accounts for a high proportion of the entire game. When players search, if there is a clear target component, the search system must accurately find the target component. If there is no clear target component, it must recommend components that meet the search intent to the player. However, the early object recommendation solutions in the game and the open source general search solutions based on the search data analysis engine (Elasticsearch) did not fully utilize the display capabilities of the search portal before and during the search, nor did they consider the game's revenue indicators, and could not improve the player experience and game revenue through personalized recommendations. To this end, the embodiment of the present application utilizes player search big data, through matching recall, offline sorting, online sorting and other algorithm stages, fully utilizes the interface recommendation search terms before and during the search, based on the player's historical search big data personalized recommendation search terms and components, and uses various natural language processing algorithms for more detailed semantic matching. When sorting, semantic relevance and revenue indicators are comprehensively considered, thereby realizing search term recommendation, input association and component search recommendation, thereby constructing a game UGC content search system that integrates recommendation algorithms and covers the entire search process. Based on the practical data indicators of the embodiment of the present application, it is shown that the embodiment of the present application not only improves the player's component search experience, allowing players to find the components they need to buy faster, but also recommends components with higher estimated revenue when the player's search target is vague, which can bring a significant increase in game revenue. After rigorous online AB testing, the search solution provided by the embodiment of the present application can bring more than 5% increase in product revenue, which can effectively improve the efficiency of object search and recommendation.

[0101] From the above, it can be seen that the embodiment of the present application receives the first keyword that has been input through the input box, determines the candidate search terms that match the first keyword; determines the historical interaction information corresponding to the candidate search terms in at least one user interaction behavior; determines the transaction feature information of the candidate search terms based on the transaction information of the first candidate objects similar to each candidate search term; sorts the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; displays the candidate search terms; in response to a keyword determination event, obtains a target keyword corresponding to the keyword determination event, and determines multiple target objects that match the target keyword; and makes recommendations based on multiple target objects. In this way, by determining the candidate search terms associated with the first keyword currently entered by the user in the input box, and then sorting the candidate search terms according to the historical interaction information and transaction feature information corresponding to the candidate search terms, the candidate search terms are recommended to the user for selection based on the sorting results. This allows the user to quickly and accurately determine the target keyword to be searched, thereby determining multiple target objects matching the target keyword for recommendation to the user, and achieving efficient and accurate recommendation of objects of interest to the user. At the same time, recommending search terms based on transaction feature information can increase the probability of users purchasing the recommended target objects, thereby increasing the benefits brought in the search scenario, and further improving the efficiency of object recommendation.

[0102] In order to better implement the above method, an embodiment of the present invention further provides an object recommendation device, which can be integrated in an electronic device, and the electronic device can be a server.

[0103] For example, Figure 5 As shown, it is a schematic diagram of the structure of the object recommendation device provided in an embodiment of the present application, and the object recommendation device may include a receiving unit 201, a first determining unit 202, a second determining unit 203, a sorting unit 204, a display unit 205, a third determining unit 206 and a recommending unit 207, as follows:

[0104] A receiving unit 201 is configured to receive a first keyword inputted through an input box, and determine a candidate search term matching the first keyword;

[0105] A first determining unit 202, configured to determine historical interaction information corresponding to a candidate search term in at least one user interaction behavior;

[0106] A second determining unit 203, configured to determine transaction feature information of the candidate search terms based on transaction information of the first candidate objects similar to each candidate search term;

[0107] A sorting unit 204, configured to sort the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term;

[0108] A display unit 205, used to display candidate search terms;

[0109] The third determination unit 206 is used to obtain a target keyword corresponding to the keyword determination event in response to the keyword determination event, and determine a plurality of target objects matching the target keyword;

[0110] The recommendation unit 207 is used to make recommendations based on multiple target objects.

[0111] In some embodiments, the transaction information includes at least one of a transaction price, a buyer's location, a transaction time, and a transaction frequency. The second determining unit 203 is configured to:

[0112] Calculating price feature information of the candidate search terms based on transaction prices of first candidate objects similar to each candidate search term;

[0113] and / or, based on the buyer locations of the first candidate objects similar to each candidate search term, determining the transaction concentration area of ​​the first candidate objects under the same candidate search term, and obtaining the transaction area feature information of the candidate search term based on the transaction concentration area;

[0114] and / or, based on the transaction time of the first candidate objects similar to each candidate search term, determining the concentrated transaction time period of the first candidate objects under the same candidate search term, and obtaining the transaction time period characteristic information of the candidate search term based on the concentrated transaction time period;

[0115] And / or, based on the transaction frequency of the first candidate object similar to each candidate search term, the transaction frequency feature information of the candidate search term is calculated.

[0116] In some embodiments, the historical interaction information includes at least one of a click rate, an add-to-cart rate, and a purchase rate.

[0117] In some embodiments, the sorting unit 204 is configured to:

[0118] Calculate the interaction score of each candidate search term based on the historical interaction information corresponding to each candidate search term;

[0119] sorting the candidate search terms based on the interaction scores, and selecting target candidate search terms from the sorted candidate search terms;

[0120] sorting the target candidate search terms based on the transaction feature information of each target candidate search term to obtain sorted target candidate search terms;

[0121] The display unit 205 is used for:

[0122] Display the ranked target candidate search terms.

[0123] In some embodiments, the recommendation unit 207 includes:

[0124] A purchase rate determination subunit, used to determine the first purchase probability of the target user for the target object;

[0125] A benefit determination subunit, used to determine first estimated benefit information of a target object based on the first purchase probability;

[0126] The object recommendation subunit is used to sort the target objects based on the estimated revenue information, and to recommend the target users based on the sorted target objects.

[0127] In some embodiments, the purchase rate determination subunit is used to:

[0128] Obtain user characteristics of the target user and object characteristics of the target object;

[0129] Based on the user characteristics, the target keywords and the object characteristics of the target object, a first purchase probability of the target user purchasing the target object is predicted.

[0130] In some embodiments, the purchase rate determination subunit is used to:

[0131] If the signal transmission quality of the target user's terminal meets the preset abnormal condition, a second purchase probability of the target object being purchased is obtained as the first purchase probability of the target user for the target object.

[0132] In some embodiments, the second purchase probability includes at least one of a probability that the target object is purchased in the search scenario, a probability that the target object is purchased in all scenarios, and a probability that a similar object corresponding to the target object is purchased.

[0133] In some embodiments, the object recommendation device further includes an offline sorting unit, which is used to:

[0134] Based on at least one matching dimension, obtaining initial objects matching a plurality of preset search terms;

[0135] Based on the priorities corresponding to the matching dimensions, the initial objects that match the search term in each matching dimension are sorted to obtain sorted initial objects;

[0136] Determine a second purchase probability and price corresponding to the initial object;

[0137] Calculating second estimated revenue information of the initial object based on the second purchase probability and the price;

[0138] Based on the second estimated revenue information, the sorted initial objects under each matching dimension are sorted to obtain candidate objects corresponding to each search term;

[0139] The third determining unit 206 is configured to:

[0140] Filter out target search terms that match the target keywords from the search terms;

[0141] Based on the candidate objects corresponding to the target search term, multiple target objects matching the target keyword are determined.

[0142] In some embodiments, the object recommendation device further includes a search term recommendation unit, which is used to:

[0143] Determining, based on the user association information of the target user, a second candidate object that the target user is interested in, and a preference score of the target user for the second candidate object;

[0144] Obtaining an associated search term associated with the second candidate object;

[0145] determining an interest score of the associated search term based on the preference score of the second candidate object associated with each associated search term;

[0146] ranking the associated search terms based on their interest scores;

[0147] Based on the ranking results of the associated search terms, search term recommendations are made through the target user's terminal.

[0148] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0149] As can be seen from the above, in the embodiment of the present application, the receiving unit 201 receives the input first keyword through the input box, and determines the candidate search terms that match the first keyword; the first determination unit 202 determines the historical interaction information corresponding to the candidate search terms in at least one user interaction behavior; the second determination unit 203 determines the transaction feature information of the candidate search terms based on the transaction information of the first candidate objects similar to each candidate search term; the sorting unit 204 sorts the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; the display unit 205 displays the candidate search terms; the third determination unit 206 responds to the keyword determination event, obtains the target keyword corresponding to the keyword determination event, and determines multiple target objects that match the target keyword; the recommendation unit 207 makes recommendations based on multiple target objects. In this way, by determining the candidate search terms associated with the first keyword currently entered by the user in the input box, and then sorting the candidate search terms according to the historical interaction information and transaction feature information corresponding to the candidate search terms, the candidate search terms are recommended to the user for selection based on the sorting results. This allows the user to quickly and accurately determine the target keyword to be searched, thereby determining multiple target objects matching the target keyword for recommendation to the user, and achieving efficient and accurate recommendation of objects of interest to the user. At the same time, recommending search terms based on transaction feature information can increase the probability of users purchasing the recommended target objects, thereby increasing the benefits brought in the search scenario, and further improving the efficiency of object recommendation.

[0150] The present application also provides an electronic device, such as Figure 6 As shown, it shows a schematic diagram of the structure of an electronic device involved in an embodiment of the present application, and the electronic device may be a server. Specifically:

[0151] The electronic device 300 includes a processor 301 having one or more processing cores, a memory 302 having one or more computer-readable storage media, and a computer program stored in the memory 302 and executable on the processor. The processor 301 is electrically connected to the memory 302. Those skilled in the art will appreciate that the electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0152] The processor 301 is the control center of the electronic device 300. It uses various interfaces and lines to connect various parts of the entire electronic device 300, executes various functions of the electronic device 300 and processes data by running or loading software programs and / or modules stored in the memory 302, and calling data stored in the memory 302, thereby monitoring the electronic device 300 as a whole.

[0153] In the embodiment of the present application, the processor 301 in the electronic device 300 will load instructions corresponding to the processes of one or more application programs into the memory 302 according to the following steps, and the processor 301 will run the application programs stored in the memory 302 to implement various functions:

[0154] receiving a first keyword inputted through an input box, and determining candidate search terms matching the first keyword;

[0155] Determining historical interaction information corresponding to the candidate search term in at least one user interaction behavior;

[0156] Determining transaction feature information of the candidate search terms based on transaction information of the first candidate objects similar to each candidate search term;

[0157] Sorting the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term;

[0158] Display candidate search terms;

[0159] In response to a keyword determination event, obtaining a target keyword corresponding to the keyword determination event, and determining a plurality of target objects matching the target keyword;

[0160] Make recommendations based on multiple target objects.

[0161] This solution can receive an input first keyword through an input box, determine candidate search terms that match the first keyword; determine historical interaction information corresponding to the candidate search terms on at least one user interaction behavior; determine transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each candidate search term; sort the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; display the candidate search terms; in response to a keyword determination event, obtain a target keyword corresponding to the keyword determination event, and determine multiple target objects that match the target keyword; and make recommendations based on multiple target objects. In this way, by determining the candidate search terms associated with the first keyword currently entered by the user in the input box, and then sorting the candidate search terms according to the historical interaction information and transaction feature information corresponding to the candidate search terms, the candidate search terms are recommended to the user for selection based on the sorting results. This allows the user to quickly and accurately determine the target keyword to be searched, thereby determining multiple target objects matching the target keyword for recommendation to the user, and achieving efficient and accurate recommendation of objects of interest to the user. At the same time, recommending search terms based on transaction feature information can increase the probability of users purchasing the recommended target objects, thereby increasing the benefits brought in the search scenario, and further improving the efficiency of object recommendation.

[0162] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0163] Optional, such as Figure 6 As shown, the electronic device 300 further includes: a touch screen 303, a radio frequency circuit 304, an audio circuit 305, an input unit 306, and a power supply 307. The processor 301 is electrically connected to the touch screen 303, the radio frequency circuit 304, the audio circuit 305, the input unit 306, and the power supply 307, respectively. Those skilled in the art can understand that Figure 6 The electronic device structure shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0164] The touch display screen 303 can be used to display a graphical user interface and receive operation instructions generated by the user acting on the graphical user interface. The touch display screen 303 may include a display panel and a touch panel. Among them, the display panel can be used to display information input by the user or information provided to the user and various graphical user interfaces of the electronic device, and these graphical user interfaces can be composed of graphics, text, icons, videos and any combination thereof. Optionally, the display panel can be configured in the form of a liquid crystal display (LCD, Liquid Crystal Display), an organic light-emitting diode (OLED, Organic Light-Emitting Diode) and the like. The touch panel can be used to collect the user's touch operation on or near it (such as the user uses any suitable object or attachment such as a finger, a stylus, etc. on the touch panel or near the touch panel), and generate corresponding operation instructions, and the operation instructions execute corresponding programs. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch direction, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into the touch point coordinates, and then sends it to the processor 301, and can receive the command sent by the processor 301 and execute it. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it is transmitted to the processor 301 to determine the type of touch event, and then the processor 301 provides a corresponding visual output on the display panel according to the type of touch event. In an embodiment of the present application, the touch panel and the display panel can be integrated into the touch display screen 303 to realize the input and output functions. However, in some embodiments, the touch panel and the touch panel can be used as two independent components to realize the input and output functions. That is, the touch display screen 303 can also be used as a part of the input unit 306 to realize the input function.

[0165] The radio frequency circuit 304 may be used to send and receive radio frequency signals, so as to establish wireless communication with a network device or other electronic devices through wireless communication, and to send and receive signals with the network device or other electronic devices.

[0166] The audio circuit 305 can be used to provide an audio interface between the user and the electronic device through a speaker and a microphone. The audio circuit 305 can transmit the electrical signal converted from the received audio data to the speaker, which is converted into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 305 and converted into audio data, and then the audio data is output to the processor 301 for processing, and then sent to another electronic device through the radio frequency circuit 304, or the audio data is output to the memory 302 for further processing. The audio circuit 305 may also include an earplug jack to provide communication between an external headset and the electronic device.

[0167] The input unit 306 may be used to receive input numbers, character information or user feature information (such as fingerprint, iris, facial information, etc.), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0168] The power supply 307 is used to supply power to various components of the electronic device 300. Optionally, the power supply 307 can be logically connected to the processor 301 through a power management system, so that the power management system can manage charging, discharging, and power consumption. The power supply 307 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0169] although Figure 6 Not shown, the electronic device 300 may also include a camera, a sensor, a wireless fidelity module, a Bluetooth module, etc., which will not be described in detail here.

[0170] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. It should be noted that the electronic device provided in the embodiment of the present application and the object recommendation method in the above embodiment belong to the same concept. The specific implementation process is detailed in the above method embodiment and will not be repeated here.

[0171] As can be seen from the above, the electronic device provided by the embodiment of the present application can receive an input first keyword through an input box, determine a candidate search term that matches the first keyword; determine historical interaction information corresponding to the candidate search term in at least one user interaction behavior; determine transaction feature information of the candidate search term based on transaction information of a first candidate object similar to each candidate search term; sort the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; display the candidate search terms; in response to a keyword determination event, obtain a target keyword corresponding to the keyword determination event, and determine multiple target objects that match the target keyword; and make recommendations based on multiple target objects. In this way, by determining the candidate search terms associated with the first keyword currently entered by the user in the input box, and then sorting the candidate search terms according to the historical interaction information and transaction feature information corresponding to the candidate search terms, the candidate search terms are recommended to the user for selection based on the sorting results. This allows the user to quickly and accurately determine the target keyword to be searched, thereby determining multiple target objects matching the target keyword for recommendation to the user, and achieving efficient and accurate recommendation of objects of interest to the user. At the same time, recommending search terms based on transaction feature information can increase the probability of users purchasing the recommended target objects, thereby increasing the benefits brought in the search scenario, and further improving the efficiency of object recommendation.

[0172] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by a computer program, or by controlling related hardware through a computer program. The computer program may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0173] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored, and the computer program can be loaded by a processor to execute the steps in any object recommendation method provided in the embodiment of the present application. For example, the computer program can execute the following steps:

[0174] receiving a first keyword inputted through an input box, and determining candidate search terms matching the first keyword;

[0175] Determining historical interaction information corresponding to the candidate search term in at least one user interaction behavior;

[0176] Determining transaction feature information of the candidate search terms based on transaction information of the first candidate objects similar to each candidate search term;

[0177] Sorting the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term;

[0178] Display candidate search terms;

[0179] In response to a keyword determination event, obtaining a target keyword corresponding to the keyword determination event, and determining a plurality of target objects matching the target keyword;

[0180] Make recommendations based on multiple target objects.

[0181] This solution can receive an input first keyword through an input box, determine candidate search terms that match the first keyword; determine historical interaction information corresponding to the candidate search terms on at least one user interaction behavior; determine transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each candidate search term; sort the candidate search terms based on the transaction feature information and historical interaction information corresponding to each candidate search term; display the candidate search terms; in response to a keyword determination event, obtain a target keyword corresponding to the keyword determination event, and determine multiple target objects that match the target keyword; and make recommendations based on multiple target objects. In this way, by determining the candidate search terms associated with the first keyword currently entered by the user in the input box, and then sorting the candidate search terms according to the historical interaction information and transaction feature information corresponding to the candidate search terms, the candidate search terms are recommended to the user for selection based on the sorting results. This allows the user to quickly and accurately determine the target keyword to be searched, thereby determining multiple target objects matching the target keyword for recommendation to the user, and achieving efficient and accurate recommendation of objects of interest to the user. At the same time, recommending search terms based on transaction feature information can increase the probability of users purchasing the recommended target objects, thereby increasing the benefits brought in the search scenario, and further improving the efficiency of object recommendation.

[0182] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0183] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0184] Since the computer program stored in the computer-readable storage medium can execute the steps in any object recommendation method provided in the embodiments of the present application, the beneficial effects that can be achieved by any object recommendation method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0185] Among them, according to one aspect of the present application, a computer program product is provided, which includes a computer program, and the computer program is stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device executes the methods provided in the various optional implementations provided in the above embodiments.

[0186] The above is a detailed introduction to an object recommendation method, device, storage medium and electronic device provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. An object recommendation method, characterized in that: include: receiving a first keyword inputted through an input box, and determining a candidate search term matching the first keyword; Determining historical interaction information corresponding to the candidate search term in at least one user interaction behavior; Determining transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each of the candidate search terms; sorting the candidate search terms based on the transaction feature information and the historical interaction information corresponding to each of the candidate search terms; displaying the candidate search terms; In response to a keyword determination event, obtaining a target keyword corresponding to the keyword determination event, and determining a plurality of target objects matching the target keyword; Recommendations are made based on the multiple target objects.

2. The object recommendation method according to claim 1, characterized in that: The transaction information includes at least one of transaction price, buyer location, transaction time, and transaction frequency; The determining of transaction feature information of the candidate search terms based on transaction information of the first candidate objects similar to each of the candidate search terms includes: Calculating price feature information of the candidate search terms based on transaction prices of first candidate objects similar to each of the candidate search terms; and / or, based on the buyer locations of the first candidate objects similar to each of the candidate search terms, determining a transaction concentration area of ​​the first candidate objects under the same candidate search terms, and obtaining transaction area feature information of the candidate search terms based on the transaction concentration area; and / or, based on the transaction time of the first candidate objects similar to each of the candidate search terms, determining the concentrated transaction time period of the first candidate objects under the same candidate search term, and obtaining the transaction time period characteristic information of the candidate search term based on the concentrated transaction time period; And / or, based on the transaction frequency of the first candidate object similar to each of the candidate search terms, the transaction frequency feature information of the candidate search terms is calculated.

3. The object recommendation method according to claim 1, characterized in that: The historical interaction information includes at least one of a click rate, an add-to-cart rate, and a purchase rate.

4. The object recommendation method according to claim 1, wherein: The sorting of the candidate search terms based on the transaction feature information and the historical interaction information corresponding to each of the candidate search terms includes: Calculating an interaction score for each of the candidate search terms based on the historical interaction information corresponding to each of the candidate search terms; sorting the candidate search terms based on the interaction scores, and selecting target candidate search terms from the sorted candidate search terms; sorting the target candidate search words based on the transaction feature information of each of the target candidate search words to obtain sorted target candidate search words; The displaying of the candidate search terms includes: The ranked target candidate search terms are displayed.

5. The object recommendation method according to claim 1, characterized in that: The making recommendations based on the multiple target objects includes: Determining a first purchase probability of the target user for the target object; Determining first estimated revenue information of the target object based on the first purchase probability; The target objects are sorted based on the estimated revenue information, and the target users are recommended based on the sorted target objects.

6. The object recommendation method according to claim 5, characterized in that: The determining a first purchase probability of the target user for the target object includes: Acquire user characteristics of the target user and object characteristics of the target object; Based on the user characteristics, the target keywords, and the object characteristics of the target object, a first purchase probability of the target user purchasing the target object is predicted.

7. The object recommendation method according to claim 5, characterized in that: The determining a first purchase probability of the target user for the target object includes: If the signal transmission quality of the terminal of the target user meets the preset abnormal condition, a second purchase probability of the target object being purchased is obtained as the first purchase probability of the target user for the target object.

8. The object recommendation method according to claim 7, characterized in that: The second purchase probability includes at least one of a probability that the target object is purchased in a search scenario, a probability that the target object is purchased in all scenarios, and a probability that a similar object corresponding to the target object is purchased.

9. The object recommendation method according to claim 5, characterized in that: Before the method of obtaining a target keyword corresponding to the keyword determination event in response to the keyword determination event and determining a plurality of target objects matching the target keyword, the method further includes: Based on at least one matching dimension, obtaining initial objects matching a plurality of preset search terms; Based on the priorities corresponding to the matching dimensions, the initial objects that match the search term on each of the matching dimensions are sorted to obtain sorted initial objects; Determining a second purchase probability and a price corresponding to the initial object; Calculating second estimated revenue information of the initial object based on the second purchase probability and the price; Based on the second estimated revenue information, the sorted initial objects under each matching dimension are sorted to obtain candidate objects corresponding to each of the search terms; The determining of a plurality of target objects matching the target keyword includes: Filtering out target search terms that match the target keyword from the search terms; Based on the candidate objects corresponding to the target search term, a plurality of target objects matching the target keyword are determined.

10. The object recommendation method according to any one of claims 1 to 9, characterized in that: Before receiving the input first keyword through the input box and determining a candidate search term matching the first keyword, the method further includes: Determining, based on the user association information of the target user, a second candidate object that the target user is interested in, and a preference score of the target user for the second candidate object; Obtaining an associated search term associated with the second candidate object; determining an interest score of the associated search term based on a preference score of the second candidate object associated with each of the associated search terms; sorting the associated search terms based on the interest scores of the associated search terms; Based on the ranking results of the associated search terms, search terms are recommended through the terminal of the target user.

11. An object recommendation device, characterized in that: include: A receiving unit, configured to receive a first keyword inputted through an input box, and determine a candidate search term matching the first keyword; A first determining unit, configured to determine historical interaction information corresponding to the candidate search term in at least one user interaction behavior; A second determining unit, configured to determine transaction feature information of the candidate search terms based on transaction information of first candidate objects similar to each of the candidate search terms; A sorting unit, configured to sort the candidate search terms based on the transaction feature information and the historical interaction information corresponding to each of the candidate search terms; A display unit, used for displaying the candidate search terms; A third determination unit, configured to, in response to a keyword determination event, obtain a target keyword corresponding to the keyword determination event, and determine a plurality of target objects matching the target keyword; A recommendation unit is used to make recommendations based on the multiple target objects.

12. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of any one of the methods of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The method comprises a computer program. When the computer program is run on an electronic device, the computer program is used to enable the electronic device to execute the steps of any one of the methods of claims 1 to 10.

Images