WO2024066253A1 - Interactive fiction-based product recommendation method and related apparatus - Google Patents

Abstract

The present application provides an interactive fiction-based product recommendation method and a related apparatus, which are applied to an electronic device in a comprehensive service system. The electronic device is provided with a voice interaction service engine; and the method comprises: calling the voice interaction service engine to obtain target voice information of a user, and performing intention recognition on the target voice information to obtain an intention recognition result of the user; according to the intention recognition result, determining a target service type adapted to the user; obtaining interaction records of the user in an interactive fiction; according to plot content, determining, from the interaction records, a target interaction record associated with the target service type; determining a target product type according to the target interaction record; determining a target product according to the target product type; obtaining product information about the target product from a server; and generating recommendation information according to the product information, and calling the voice interaction service engine to output the recommendation information. In this way, products can be recommended to users in a targeted manner, and the rates of adoption of the recommended products by the users are improved.

Description

Product recommendation method and related device based on interactive novel

This application claims the priority of the Chinese patent application filed with the China Patent Office on September 29, 2022, with application number 2022111947141 and application name “Product recommendation method and related device based on interactive novels”, the entire contents of which are incorporated by reference in this application.

Technical Field

The present application belongs to the general data processing technology field of the Internet industry, and specifically relates to a product recommendation method and related devices based on interactive novels.

Background technique

Currently, when recommending products to users through human-computer dialogue, it is based on simply searching for the corresponding product content based on the recognition results after recognizing the user's input voice, and then recommending to the user based on the search results. This product recommendation method cannot make personalized recommendations for different users, resulting in a low adoption rate after users receive the recommended results. The existing personalized recommendation method requires obtaining a large amount of users' private data in daily life, resulting in a low user experience.

Summary of the invention

The embodiment of the present application provides a product recommendation method and related devices based on interactive novels, so as to recommend products to users in a targeted manner during human-computer dialogue, thereby improving the user's adoption rate of the recommended products and the user's usage experience.

In a first aspect, an embodiment of the present application provides a product recommendation method based on interactive novels, which is applied to an electronic device in an integrated service system, wherein the integrated service system includes the electronic device and a server, the electronic device is communicatively connected with the server, and the electronic device is provided with a voice interactive service engine; the method includes:

Calling the voice interaction service engine to obtain the user's target voice information, performing intent recognition on the target voice information, and obtaining the user's intent recognition result, wherein the intent recognition result is used to characterize the user's demand intention;

Determine a target service type adapted to the user according to the intention recognition result, the target service type being a service type of at least one of the following service types supported by the server of the integrated service system: film and television recommendation service, food recommendation service, and scenic spot recommendation service;

Acquire an interaction record of the user in the interactive novel, wherein the interaction record includes interactive voice information input by the user and plot content of the interactive novel corresponding to the interactive voice information;

Determining a target interaction record associated with the target service type from the interaction record according to the plot content;

Determine a target product type according to the target interaction record, the target product type being at least one product type among a plurality of product types corresponding to the target service type, the product type being used to indicate a classification of products included in the service corresponding to the target service type;

Determine a target product according to the target product type, wherein the target product is a product belonging to the target product type;

Acquire product information about the target product from the server;

Generate recommendation information based on the product information, and call the voice interaction service engine to output the recommendation information.

In a second aspect, an embodiment of the present application provides a product recommendation device based on interactive novels, which is applied to an electronic device in an integrated service system, wherein the integrated service system includes the electronic device and a server, the electronic device is communicatively connected with the server, and the electronic device is provided with a voice interactive service engine; the device includes:

A first acquisition unit is used to call the voice interaction service engine to acquire the user's target voice information, perform intent recognition on the target voice information, and obtain the user's intent recognition result, wherein the intent recognition result is used to characterize the user's demand intention;

A first determining unit is used to determine a target service type adapted to the user according to the intention recognition result, wherein the target service type is a service type of at least one of the following service types supported by the server of the integrated service system: a film and television recommendation service, a food recommendation service, and a scenic spot recommendation service;

A second acquisition unit is used to acquire the interaction record of the user in the interactive novel, wherein the interaction record includes the interactive voice information input by the user and the plot content of the interactive novel corresponding to the interactive voice information;

A second determining unit, configured to determine a target interaction record associated with the target service type from the interaction record according to the plot content;

A third determining unit is used to determine a target product type according to the target interaction record, wherein the target product type is at least one product type among a plurality of product types corresponding to the target service type, and the product type is used to indicate a classification of products included in the service corresponding to the target service type;

A fourth determining unit, configured to determine a target product according to the target product type, wherein the target product is a product belonging to the target product type;

a third acquiring unit, configured to acquire product information about the target product from the server;

An output unit is used to generate recommendation information based on the product information and call the voice interaction service engine to output the recommendation information.

In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

In a fourth aspect, an embodiment of the present application provides a computer storage medium, characterized in that it stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute some or all of the steps described in the first aspect of this embodiment.

In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

It can be seen that in this example, the electronic device first calls the voice interaction service engine to obtain the user's target voice information, performs intent recognition on the target voice information, and obtains the user's intent recognition result, the intent recognition result is used to characterize the user's demand intention, and then determines the target service type adapted to the user according to the intent recognition result, the target service type is a service type of at least one of the following service types supported by the server of the comprehensive service system: film and television recommendation service, food recommendation service, scenic spot recommendation service, and then obtains the user's interactive novel in the interactive novel. The target service type is determined from the target interaction record, wherein the interaction record includes the interactive voice information input by the user and the plot content of the interactive novel corresponding to the interactive voice information. Then, according to the plot content, the target interaction record associated with the target service type is determined from the interaction record. Then, according to the target interaction record, the target product type is determined. The target product type is at least one product type among the multiple product types corresponding to the target service type. The product type is used to indicate the classification of the products included in the service corresponding to the target service type. Then, according to the target product type, the target product is determined. The target product is a product belonging to the target product type. Then, product information about the target product is obtained from the server. Finally, recommendation information is generated according to the product information, and the voice interaction service engine is called to output the recommendation information. In this way, through the user's interaction record for the same service type in the interactive novel, the user's preference for the product type included in the service type can be obtained, and then the user can be targeted to recommend products based on this, which can improve the user's satisfaction with the product recommendation function and improve the user's adoption rate of the recommended products. Moreover, when determining the user's preference for product types, this solution is only based on the user's interaction records in the interactive novel, without obtaining a large amount of user privacy data in daily life. Therefore, the user will not feel that they are being monitored by electronic devices, thereby improving the user's usage experience and peace of mind.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

FIG1 is a schematic diagram of the composition of an integrated service system provided in an embodiment of the present application;

FIG2 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

FIG3 is a flow chart of a product recommendation method based on interactive novels provided in an embodiment of the present application;

FIG4 is a schematic diagram of a product recommendation interface provided in an embodiment of the present application;

FIG5 is a block diagram of functional units of a product recommendation device based on interactive novels provided in an embodiment of the present application;

FIG6 is a block diagram of the functional units of another product recommendation device based on interactive novels provided in an embodiment of the present application.

Detailed ways

In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings 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 ordinary technicians in this field without creative work are within the scope of protection of this application.

The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

At present, when recommending products to users, it is not possible to make personalized recommendations to users in a targeted manner, so that the recommended services cannot meet the needs of users well, resulting in a low adoption rate of recommended products. The current solutions for personalized recommendations to users all require the acquisition of a large amount of privacy data in users' daily lives, resulting in a poor user experience.

In response to the above problems, an embodiment of the present application provides a product recommendation method and related devices based on interactive novels. The embodiment of the present application is described in detail below in conjunction with the accompanying drawings.

Please refer to Figure 1, which is a schematic diagram of the composition of a comprehensive service system provided in an embodiment of the present application. As shown in Figure 1, the comprehensive service system 10 includes an electronic device 101 and a server 102, the electronic device 101 is communicatively connected to the server 102, and the electronic device 101 is provided with a voice interaction service engine 111. The voice interaction service engine 111 can perform intent recognition on the voice information input by the user, so that the electronic device 101 can determine the product that meets the user's needs. The server 102 can support multiple services, so that when the service required by the user is a service type supported by the server 102, the electronic device 101 can recommend products to the user in combination with the product information in the server 102.

Please refer to Figure 2, which is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. As shown in Figure 2, the electronic device 101 includes a processor 120, a memory 130, a communication interface 140, and one or more programs 131, wherein the one or more programs 131 are stored in the above-mentioned memory 130 and are configured to be executed by the above-mentioned processor 120, and the one or more programs 131 include instructions for executing any step in the following method embodiment. In a specific implementation, the processor 120 is used to execute any step performed by the electronic device in the following method embodiment, and when performing data transmission such as sending, the communication interface 140 can be selectively called to complete the corresponding operation.

The electronic device involved in the embodiments of the present application may be an electronic device with communication capabilities, which may include various handheld devices with wireless communication functions, vehicle-mounted devices, wearable devices, computing devices or other processing devices connected to a wireless modem, as well as various forms of user equipment (UE), mobile stations (MS), terminal devices, etc.

Please refer to Figure 3, which is a flow chart of a product recommendation method based on interactive novels provided in an embodiment of the present application. The product recommendation method based on interactive novels is applied to an electronic device in an integrated service system, the integrated service system includes the electronic device and a server, the electronic device is connected to the server in communication, and the electronic device is provided with a voice interactive service engine. As shown in Figure 3, the method includes the following steps:

S201, calling the voice interaction service engine to obtain the user's target voice information, performing intent recognition on the target voice information, and obtaining the user's intent recognition result.

Wherein, the intention recognition result is used to characterize the user's demand intention. The intention recognition of the target voice information includes: converting the target voice information into target text information, generating a parsing graph according to the target text information, the parsing graph includes multiple parsing nodes and parsing node relationship lines, the parsing node relationship line is used to indicate the grammatical or semantic relationship between two parsing nodes, the two parsing nodes can be two adjacent parsing nodes, or two non-adjacent parsing nodes, the same parsing node can correspond to multiple parsing node relationship lines, and the parsing node is a word in the target file information. Match the parsing graph with a pre-generated knowledge graph to determine which parsing nodes can match it. A matching knowledge subgraph is provided, wherein a plurality of knowledge subgraphs constitute the knowledge graph, each knowledge subgraph includes a plurality of knowledge nodes and a relationship line between each knowledge node, a knowledge node is a word set, and a knowledge node relationship line is used to indicate a grammatical or semantic relationship between two knowledge nodes, wherein the two knowledge nodes may be two adjacent knowledge nodes or two non-adjacent knowledge nodes, and the same knowledge node may correspond to multiple knowledge node relationship lines. If there is a matching knowledge subgraph, the user intent is determined based on the triples included in the knowledge subgraph, and the included triples may be used to characterize the intent recognition result, and each triple includes two entities and an association relationship between the two entities, and the two entities may correspond to a word respectively, and the association relationship may be a semantic relationship and/or a grammatical relationship.

S202: Determine a target service type suitable for the user according to the intention recognition result.

The target service type is at least one of the following service types supported by the server of the integrated service system: film and television recommendation service, food recommendation service, and scenic spot recommendation service. The film and television recommendation service can be used to recommend TV series, movies, animations and other film and television content to users, the food recommendation service can recommend restaurants, takeaway shops, etc. to users, and the scenic spot recommendation service can be used to recommend scenic spots, routes, travel methods, etc. to users, and the travel method recommendation can also include ticket booking services.

S203, obtaining the user's interaction record in the interactive novel.

The interactive record includes the interactive voice information input by the user and the plot content of the interactive novel corresponding to the interactive voice information. The plot content can be used to indicate the plot of the novel in which the interactive record is currently located, for example, the current interactive record corresponds to the plot of the protagonist having a dinner party, or the plot of the male and female protagonists watching a movie, etc. The interactive voice information is the voice information about the plot content input by the user according to the interactive plot when using the interactive novel.

S204: Determine a target interaction record associated with the target service type from the interaction records according to the plot content.

Wherein, the determining of the interaction record associated with the target service type includes: performing text analysis on the interactive voice information input by the user included in the acquired interactive record to obtain interactive text information; determining the keyword corresponding to the target service type, and determining the interactive text information including the keyword from the acquired interactive text information as the target interactive text information; and determining the interactive record corresponding to the interactive voice information corresponding to the target interactive text information as the target interactive record. And the machine output sentence in the plot content included in the interactive record can also be obtained for text analysis to determine whether the machine output sentence includes the keyword; if it does, then determine the user input voice information corresponding to the machine output sentence; perform intent recognition on the user input voice information corresponding to the machine output sentence to obtain a recognition result, and determine whether the interactive record corresponding to the machine output sentence is the target interactive record according to the recognition result. The recognition result can indicate whether the user has a positive intention or a negative intention. If the user has a positive intention, then it is the target interactive record. For example, the machine output sentence is "Shall we go eat barbecue?" It can be seen that the machine output sentence includes the word "barbecue" as the keyword corresponding to the food recommendation service. Therefore, if the user has a positive intention, that is, the voice information input by the user in the interactive novel is "OK", then this interaction record can be determined as the target interaction record.

S205: Determine a target product type according to the target interaction record.

The target product type is at least one of the multiple product types corresponding to the target service type, and the product type is used to indicate the classification of products included in the service corresponding to the target service type. For example, the product types corresponding to the film and television recommendation service include: Xianxia, suspense, campus, etc., the food recommendation service corresponds to barbecue, hot pot, Japanese food, etc., and the scenic spot recommendation service corresponds to travel, humanities, and scenery, etc. Various types.

S206: Determine a target product according to the target product type.

The target product is a product belonging to the target product type. The target product may include multiple products. For example, when the service type is food recommendation and the target product type is hot pot type, the target product may be multiple hot pot restaurants such as hot pot restaurant 1 and hot pot restaurant 2.

S207: Acquire product information about the target product from the server.

The product information includes link information, and the user can jump to the corresponding product interface according to the link information to obtain more complete product information. When obtaining product information from the service, after the electronic device determines the target product, it can send request information for the product information corresponding to the target product to the server, and the server sends response information to the electronic device according to the request information, and the response information includes product information about the target product. Alternatively, the server first sends the product information of each product to the electronic device, and the electronic device stores the product information of each product in a hierarchical manner according to the product type and service type. Before the electronic device determines the target product, it can synchronously determine the storage path step by step according to the determined service type and product type, and then after the electronic device determines the target product, it can directly find the product information related to the target product through the storage path corresponding to the target product.

S208, generating recommendation information according to the product information, and calling the voice interaction service engine to output the recommendation information.

The recommendation information includes a machine sentence and the corresponding product information of the target product. The recommendation sentence recommends the target product to the user, and the product information is used for the user to understand the specific content of the target product.

It can be seen that in this example, the electronic device first calls the voice interaction service engine to obtain the user's target voice information, performs intent recognition on the target voice information, and obtains the user's intention recognition result, which is used to characterize the user's demand intention, and then determines the target service type adapted to the user based on the intention recognition result, and the target service type is a service type of at least one of the following service types supported by the server of the comprehensive service system: film and television recommendation service, food recommendation service, and scenic spot recommendation service, and then obtains the user's interaction record in the interactive novel, and the interaction record includes the interactive voice information input by the user and the plot content of the interactive novel corresponding to the interactive voice information. Then, according to the plot content, the target interaction record associated with the target service type is determined from the interaction record, and then the target product type is determined according to the target interaction record, the target product type is at least one product type among the multiple product types corresponding to the target service type, and the product type is used to indicate the classification of the products included in the service corresponding to the target service type, and then the target product is determined according to the target product type, and the target product is a product belonging to the target product type, and then the product information about the target product from the server is obtained, and finally the recommendation information is generated according to the product information, and the voice interaction service engine is called to output the recommendation information. In this way, through the user's interaction record for the same service type in the interactive novel, the user's preference for the product type included in the service type can be obtained, and then the product recommendation can be made to the user in a targeted manner, which can improve the user's satisfaction with the product recommendation function and improve the user's adoption rate of the recommended product. Moreover, when determining the user's preference for the product type, this solution is only determined based on the user's interaction record in the interactive novel, and does not need to obtain a large amount of privacy data of the user in daily life, so it will not make the user feel monitored by the electronic device, and improve the user's use experience and peace of mind.

In a possible example, the determining of the interaction record with the target service according to the plot content The target interaction record associated with a type includes: determining the target plot corresponding to the target service type; determining the interaction record whose plot content in the interaction record conforms to the target plot as the target interaction record; in the case where the target interaction record includes multiple, determining the target product type based on the target interaction record, including: determining the product type included in each target interaction record based on the interactive voice information included in the target interaction record; determining the product type corresponding to the majority of all product types included in multiple target interaction records as the target product type.

Among them, the product type corresponding to the target service type in each target interaction record can be determined according to the target interaction record. The determination of the product type can be determined according to the voice information input by the user in the target interaction record, or it can be determined by first determining the candidate product type according to the machine output sentence corresponding to the plot content included in the target interaction record, and then determining the final product type from the candidate product type according to the user's reply voice information to the machine output sentence. For example, the machine output sentence includes "Which do you prefer, hot pot or barbecue?", so the obtained candidate product types include two product types, "hot pot" and "barbecue". If the user answers "the second one", it can be known that the product type included in the target interaction record is "barbecue". Of course, if it is determined that the user has a negative intention for the candidate product in combination with the machine output sentence and the user's voice input information corresponding to the machine output sentence, the corresponding target interaction record can be deleted.

In the specific implementation, since there are multiple interaction records, there may be multiple product types. Of course, it is also possible that multiple interaction records correspond to the same product type. If they all correspond to the same product type, the corresponding same product type is determined as the target product type. If not, it is necessary to determine the product type with the most occurrences among the multiple product types as the target product type, that is, the product type corresponding to the majority.

It can be seen that in this example, the target interaction records are first screened out from the acquired interaction records according to the target plot corresponding to the target service type, and then the product type with the largest number of occurrences among the product types corresponding to the target interaction records is determined as the target product type. Since the product types included in the target interaction records are all product types actively selected by users in the interactive novel, the user's preference for each product type in the target service type can be determined based on the interaction record, and the product type corresponding to the majority is determined as the target product type, which can best fit the user's preferences and ensure that the final recommended product is adopted by the user.

In a possible instance, determining that the product type corresponding to the majority of all product types included in a plurality of target interaction records is the target product type includes: determining the interaction type corresponding to each target interaction record, the interaction type including selection-type interaction and question-and-answer-type interaction; determining that the product type included in the target interaction record corresponding to the selection-type interaction is the first product type, and the product type included in the target interaction record corresponding to the question-and-answer-type interaction is the second product type, to obtain a first product type set and a second product type set; determining the total number of the first product type included in the first product type set, and the total number of the second product type included in the second product type set; determining the number of the first reference product type in the first product type set, and the number of the second reference product type in the second product type set the number of second reference product types in the set, the first reference product type is the product type corresponding to the majority in the set of the first product types, and the second reference product type is the product type corresponding to the majority in the set of the second product types; determining the first quantity ratio according to the total number of the first product types and the number of the first reference product types; determining the second quantity ratio according to the total number of the second product types and the number of the second reference product types; determining the weight ratio corresponding to the selection interaction and the question-and-answer interaction, the weight of the question-and-answer interaction is greater than that of the selection interaction; calculating the scores of the first reference product type and the second reference product type according to the first quantity ratio, the second quantity ratio and the weight ratio; determining the first reference product type and the second reference product type The product type with a high score is the target product type.

Among them, the voice information input by the user included in the selection type interaction record is the selection of the product type included in the machine output sentence, while the voice information input by the user included in the question-and-answer type interaction record is the answer to the machine output sentence, and the machine output sentence does not include the product type. For example, the machine output sentence included in the interaction record is "Do you like hot pot or barbecue?", and the voice information input by the user is "barbecue", then the interaction record is a selection type interaction record. If the machine output sentence included in the interaction record is "What shall we eat for dinner?", and the voice information input by the user is "barbecue", then the interaction record is a question-and-answer type interaction record. Since the corresponding product type in the question-and-answer type interaction record is a more free choice for the user and can also be more in line with the user's product type preferences, the corresponding weight of the question-and-answer type interaction record is higher than that of the selection type interaction record.

In a specific implementation, after obtaining multiple interaction records, different product types can be classified according to the interaction type to obtain two product type sets. It should be noted that the first product type set and the second product type set can include the same product types. Then determine the most product types included in each product type in the two product type sets, as well as the number of the most product types. For example, the product types included in the first product type set are "hot pot", "barbecue" and "Japanese food", among which "hot pot" appears 3 times, "barbecue" appears 5 times, and "Japanese food" appears 2 times. Therefore, the first reference product type corresponding to the majority in the first product type set is "barbecue", with a quantity of 5. The total number of the first product type set is 10, and the corresponding first quantity accounts for 5÷10=0.5. The parameters in the second product type set are determined in the same way. If the second reference type in the determined second product type set is "grilled fish" and the second quantity accounts for 0.4, then since the weight of the selection interaction is greater than the weight of the question-and-answer interaction, if the weight ratio of the selection interaction to the question-and-answer interaction is 2:1, then it can be seen that the score of the first reference product type is 0.5 and the score of the second reference product type is 0.8. Therefore, it can be determined that the target product type is the second reference product type, i.e., "grilled fish".

It can be seen that in this example, the user's preferences are comprehensively determined based on the interaction type corresponding to the target interaction record to obtain the target product type. This can be closer to the user's actual product type preferences, reduce the restrictions of the interactive novel framework on the user's product type preferences, and increase the adoption rate of the products ultimately recommended to the user.

In a possible example, determining the product type with a higher score between the first reference product type and the second reference product type as the target product type includes: determining the product type with a higher score as the alternative product type; obtaining historical recommendation records, the historical recommendation records being recommendation records for products corresponding to the target plot; obtaining the recommendation adoption status corresponding to each historical recommendation record; determining whether the alternative product type is included in the historical recommendation records; if so, determining the recommendation adoption rate of the alternative product type based on the recommendation adoption status; and determining the alternative product type as the target product type when the recommendation adoption rate is higher than a preset adoption rate.

Among them, even if the product type with the highest score, that is, the alternative product type, has been determined, it is also necessary to determine the user's adoption rate of the alternative product type, because in real life, whether a user adopts a recommended product is not only related to the user's preferences, but also to many objective factors. For example, the user likes to eat barbecue, but there is no barbecue restaurant around the user's location. Therefore, even if the target product of the product type "barbecue" is recommended according to the user's preferences, the user will still not adopt it. Determining whether the user adopts the recommended product type may include: after outputting the product corresponding to the alternative product type, determining whether the user performs a confirmation operation on the product within a preset time period. If so, it indicates that the recommendation of the product type is adopted, otherwise it is not adopted. The confirmation operation includes the user replying to a voice message with confirmation intention. The adoption rate is the ratio of the number of times an alternative product type is adopted to the number of times the alternative product type is recommended.

It can be seen that in this example, combining the adoption rates of alternative product types to comprehensively judge the target product type can make the final recommended target product fit the user's actual situation, making the recommended product more likely to be adopted by the user, and improving the user's trust in the product recommendation service.

In a possible instance, the method further includes: when the recommendation adoption rate is not higher than the preset adoption rate, determining whether the product type included in the most recent target interaction record is the same as the alternative product type; if they are the same, determining the product type with the highest recommendation adoption rate based on the recommendation adoption situation, and determining the alternative product type and the product type with the highest recommendation adoption rate as the target product type; if they are not the same, determining the alternative product type and the product type included in the most recent target interaction record as the target product type.

The most recent target interaction record is the interaction record in the target interaction record whose interaction time is closest to the current time. If the product type included in the most recent interaction record is the candidate product type, it means that the user's current preference for the candidate product type is still high, but because the user's adoption rate for the candidate product type is low, the product type with a higher user adoption rate can also be recommended to the user. If the product type included in the most recent interaction record is not the candidate product type, when making product recommendations, products in the product type included in the most recent interaction record can also be recommended at the same time.

It can be seen that in this example, if the recommendation adoption rate of an alternative product type is not high, then another product type recommended simultaneously with the alternative product type is determined based on the product type included in the most recent interaction record. This can give users more choices, and the recommended products are all products that users prefer, which can improve the user experience.

In a possible instance, generating recommendation information based on the product information includes: determining, from the target interaction record, that the interaction record corresponding to the target product type is the final interaction record; obtaining final plot information included in the final interaction record, wherein the final plot information includes character information and time information; generating a first recommendation statement based on the character information, the time information and the target product type; generating a second recommendation statement and recommended content based on the target product; generating a final recommendation statement based on the first recommendation statement and the second recommendation statement; and generating the recommendation information based on the final recommendation statement and the product information.

Wherein, when there are multiple target interaction records, the target interaction record generated last is determined as the final interaction record. The character information is the character information associated with the target service type that appears in the final plot information. When the character information includes multiple characters, the character that is most closely related to the role played by the user in the interactive novel among the multiple characters is determined as the character information included in the final plot information. The method for determining the time information may include determining the time when the target interaction record is generated. If the time interval between the time and the current time is less than a preset interval, the time information is determined to be the time when the target interaction record is generated. If the time interval is not less than the preset interval, the event related to the target service type included in the final plot information is determined, and the time information is characterized by the event related to the service type. Please refer to Figure 4, which is a schematic diagram of a product recommendation interface provided by an embodiment of the present application. As shown in Figure 4, user A sends a target voice message to the voice interaction server engine (named "Xiaowu" in the figure), and the target voice message is "What should I eat for dinner today?" Then the voice interaction service engine can perform intent recognition on the target voice message to determine that the target service type is a food recommendation service. When it is determined that the target product is xxxxx barbecue restaurant, when sending the product information of the barbecue restaurant to the user, it is necessary to output a recommendation sentence at the same time. The recommendation sentence can be generated based on the character information and time information in the final interaction record. As shown in Figure 4, since the time interval is only one day, which is less than the preset The time information at this time is represented by the generation time of the final interaction record. The final recommendation sentence recommends products to users and evokes users' memories of interactive novels. The product information is used for users to gain a deeper understanding of the recommended products.

It can be seen that in this example, the recommendation sentences are generated by combining the character information related to the novel content in the interaction record, which can not only evoke users' memories of interactive novels and increase the adoption rate of recommended products, but also enable users to use interactive novels again, improve the reading rate of interactive novels, and make users feel more immersed when reading interactive novels again.

In a possible example, determining the target product according to the target product type includes: determining user information according to the target service type; determining alternative products according to the target product type and the user information; determining whether the target service type is suitable for repeated recommendations for the same product; if so, determining the target product from the alternative products; if not, deleting the recommended products from the alternative products, and determining the target product from the remaining alternative products.

Among them, the user information corresponding to different service types may not be exactly the same. The user information corresponding to the film and television recommendation service may include the user's age information. Since some film and television works have viewing age restrictions, such products that are not suitable for users to watch need to be screened out when determining the alternative products. The user information corresponding to the food recommendation service may include the user's current location, so when determining the alternative products, restaurants that are greater than the preset distance from the user's current location can be screened out. The user information corresponding to the scenic spot recommendation service may be the user's disposable time and current location, and the play time of the corresponding scenic spot in the alternative product needs to be within the disposable time. And because some target services are not suitable for repeated recommendations for the same product, while others are suitable, for example, for the film and television recommendation service, it is not suitable to repeatedly recommend the same film and television work. If a film and television work has been recommended, it needs to be deleted from the alternative products. For the food recommendation service, multiple recommendations can be made for the same restaurant, and there is no need to delete the recommended product from the alternative products.

It can be seen that in this example, the target product is comprehensively determined based on the user information and whether the target service type is suitable for repeated recommendation, so that the recommended target product meets the user's needs and at the same time improves the user's satisfaction and usage experience.

Consistent with the above-mentioned embodiment, please refer to Figure 5, which is a functional unit composition block diagram of a product recommendation device based on interactive novels provided in an embodiment of the present application, which is applied to an electronic device in an integrated service system, wherein the integrated service system includes the electronic device and a server, the electronic device is communicatively connected to the server, and the electronic device is provided with a voice interaction service engine; the product recommendation device 40 based on interactive novels includes: a first acquisition unit 401, which is used to call the voice interaction service engine to obtain the user's target voice information, perform intent recognition on the target voice information, and obtain the user's intent recognition result, wherein the intent recognition result is used to characterize the user's demand intention; a first determination unit 402, which is used to determine a target service type adapted to the user according to the intent recognition result, wherein the target service type is a service type of at least one of the following service types supported by the server of the integrated service system: film and television recommendation service , food recommendation service, scenic spot recommendation service; a second acquisition unit 403, used to obtain the interaction record of the user in the interactive novel, the interaction record includes the interactive voice information input by the user and the plot content of the interactive novel corresponding to the interactive voice information; a second determination unit 404, used to determine the target interaction record associated with the target service type from the interaction record according to the plot content; a third determination unit 405, used to determine the target product type according to the target interaction record, the target product type is at least one product type among the multiple product types corresponding to the target service type, and the product type is used to indicate the classification of products included in the service corresponding to the target service type; a fourth determination unit 406, used to determine the target product according to the target product type, the target product is a product belonging to the target product type; a third acquisition unit 407, used to obtain information about the target service type from the server. output unit 408, used to generate recommendation information according to the product information, and call the voice interaction service engine to output the recommendation information.

In a possible example, in determining the target interaction record associated with the target service type from the interaction record according to the plot content, the third determination unit 405 is specifically used to: determine the target plot corresponding to the target service type; determine the interaction record whose plot content in the interaction record conforms to the target plot as the target interaction record; in the case where the target interaction record includes multiple target interaction records, in determining the target product type according to the target interaction record, the fourth determination unit 406 is specifically used to: determine the product type included in each target interaction record according to the interactive voice information included in the target interaction record; determine the product type corresponding to the majority of all product types included in multiple target interaction records as the target product type.

In a possible example, in terms of determining that the product type corresponding to the majority of all product types included in a plurality of target interaction records is the target product type, the fourth determination unit 406 is specifically used to: determine the interaction type corresponding to each target interaction record, the interaction type including selection-type interaction and question-and-answer-type interaction; determine that the product type included in the target interaction record corresponding to the selection-type interaction is the first product type, and the product type included in the target interaction record corresponding to the question-and-answer-type interaction is the second product type, to obtain a first product type set and a second product type set; determine the total number of the first product types included in the first product type set, and the total number of the second product types included in the second product type set; determine the number of the first reference product type in the first product type set, and the number of the second reference product type in the second product type set. the number of second reference product types, the first reference product type being the product type corresponding to the majority in the first product type set, and the second reference product type being the product type corresponding to the majority in the second product type set; determining a first quantity ratio according to the total number of the first product type and the number of the first reference product type; determining a second quantity ratio according to the total number of the second product type and the number of the second reference product type; determining a weight ratio corresponding to the selection interaction and the question-and-answer interaction, the weight of the question-and-answer interaction being greater than that of the selection interaction; calculating scores of the first reference product type and the second reference product type according to the first quantity ratio, the second quantity ratio and the weight ratio; determining the product type with a higher score between the first reference product type and the second reference product type as the target product type.

In a possible example, in terms of determining that the product type with a high score among the first reference product type and the second reference product type is the target product type, the fourth determination unit 406 is specifically used to: determine that the product type with a high score is the alternative product type; obtain historical recommendation records, which are recommendation records for products corresponding to the target plot; obtain the recommendation adoption status corresponding to each historical recommendation record; determine whether the historical recommendation records include the alternative product type; if so, determine the recommendation adoption rate of the alternative product type based on the recommendation adoption status; and when the recommendation adoption rate is higher than a preset adoption rate, determine that the alternative product type is the target product type.

In a possible example, the fourth determination unit 406 is specifically used to: when the recommendation adoption rate is not higher than the preset adoption rate, determine whether the product type included in the most recent target interaction record is the same as the alternative product type; if they are the same, determine the product type with the highest recommendation adoption rate based on the recommendation adoption situation, and determine the alternative product type and the product type with the highest recommendation adoption rate as the target product type; if they are not the same, determine the alternative product type and the product type included in the most recent target interaction record as the target product type.

In a possible example, in generating recommendation information according to the product information, the output unit 408 Specifically used for: determining from the target interaction record that the interaction record corresponding to the target product type is the final interaction record; obtaining the final plot information included in the final interaction record, the final plot information including character information and time information; generating a first recommendation statement based on the character information, the time information and the target product type; generating a second recommendation statement and recommended content based on the target product; generating a final recommendation statement based on the first recommendation statement and the second recommendation statement; generating the recommendation information based on the final recommendation statement and the product information.

In a possible example, in terms of determining the target product according to the target product type, the output unit 408 is specifically used to: determine user information according to the target service type; determine alternative products according to the target product type and the user information; determine whether the target service type is suitable for repeated recommendations for the same product; if so, determine the target product from the alternative products; if not, delete the recommended products from the alternative products, and determine the target product from the remaining alternative products.

It can be understood that since the method embodiment and the device embodiment are different presentation forms of the same technical concept, the content of the method embodiment part in this application should be synchronously adapted to the device embodiment part and will not be repeated here.

In the case of using an integrated unit, please refer to Figure 6, which is a functional unit composition block diagram of another product recommendation device based on interactive novels provided in an embodiment of the present application. In Figure 6, the product recommendation device 500 based on interactive novels includes: a processing module 512 and a communication module 511. The processing module 512 is used to control and manage the actions of the product recommendation device based on patch advertisements, for example, to execute the steps of the first acquisition unit 401, the first determination unit 402, the second acquisition unit 403, the second determination unit 404, the third determination unit 405, the fourth determination unit 406, the third acquisition unit 407 and the output unit 408, and/or to perform other processes of the technology described herein. The communication module 511 is used for interaction between the product recommendation device based on interactive novels and other devices. As shown in Figure 6, the product recommendation device based on interactive novels may also include a storage module 513, which is used to store program code and data of the product recommendation device based on interactive novels.

Among them, the processing module 412 can be a processor or a controller, for example, it can be a central processing unit (CPU), a general processor, a digital signal processor (DSP), an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logic boxes, modules and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of DSP and microprocessor, etc. The communication module 411 can be a transceiver, an RF circuit or a communication interface, etc. The storage module 413 can be a memory.

All relevant contents of each scenario involved in the above method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here. The above interactive novel-based product recommendation device 500 can execute the interactive novel-based product recommendation method shown in FIG. 3 .

The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that, in order to realize the above functions, the electronic device includes a hardware structure and a software module corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

The embodiment of the present application can divide the functional units of the electronic device according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

An embodiment of the present application also provides a chip, wherein the chip includes a processor for calling and running a computer program from a memory, so that a device equipped with the chip executes some or all of the steps described for the electronic device in the above method embodiment.

An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiments, and the above computer includes an electronic device.

The present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any method described in the method embodiment. The computer program product may be a software installation package, and the computer includes an electronic device.

It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including several instructions for enabling a computer device (which may be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the above methods in various embodiments of the present application. The aforementioned memory includes: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable memory, and the memory may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

The embodiments of the present application are introduced in detail above. 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 and core idea of the present application. At the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present application, and can make various changes and modifications, including the combination of the above-mentioned different functions and implementation steps, including software and hardware implementation methods, all of which are within the scope of protection of the present application.

Claims (10)

1. A product recommendation method based on interactive novels, characterized in that it is applied to an electronic device in an integrated service system, the integrated service system includes the electronic device and a server, the electronic device is communicatively connected with the server, and the electronic device is provided with a voice interactive service engine; the method includes:

   Calling the voice interaction service engine to obtain the user's target voice information, performing intent recognition on the target voice information, and obtaining the user's intent recognition result, wherein the intent recognition result is used to characterize the user's demand intention;

   Determine a target service type adapted to the user according to the intention recognition result, the target service type being a service type of at least one of the following service types supported by the server of the integrated service system: film and television recommendation service, food recommendation service, and scenic spot recommendation service;

   Acquire an interaction record of the user in the interactive novel, wherein the interaction record includes interactive voice information input by the user and plot content of the interactive novel corresponding to the interactive voice information;

   Determining a target interaction record associated with the target service type from the interaction record according to the plot content;

   Determine a target product type according to the target interaction record, the target product type being at least one product type among a plurality of product types corresponding to the target service type, the product type being used to indicate a classification of products included in the service corresponding to the target service type;

   Determine a target product according to the target product type, wherein the target product is a product belonging to the target product type;

   Acquire product information about the target product from the server;

   Generate recommendation information based on the product information, and call the voice interaction service engine to output the recommendation information.
2. The method according to claim 1, characterized in that the step of determining the target interaction record associated with the target service type from the interaction record according to the plot content comprises:

   Determine a target plot corresponding to the target service type;

   Determine the interaction record whose plot content in the interaction record matches the target plot as the target interaction record;

   In the case that the target interaction record includes multiple ones, determining the target product type according to the target interaction record includes:

   Determining the product type included in each target interaction record according to the interactive voice information included in the target interaction record;

   The product type corresponding to the majority among all product types included in the multiple target interaction records is determined as the target product type.
3. The method according to claim 2, characterized in that determining the product type corresponding to the majority of all product types included in the plurality of target interaction records as the target product type comprises:

   Determine the interaction type corresponding to each target interaction record, where the interaction type includes selection interaction and question-and-answer interaction;

   Determine that the product type included in the target interaction record corresponding to the selection interaction is the first product type, and the product type included in the target interaction record corresponding to the question-and-answer interaction is the second product type, to obtain a first product type set and a second product type set;

   Determine the total quantity of the first product type included in the first product type set, and the total quantity of the second product type The total quantity of the second product type included in the type set;

   Determine the number of first reference product types in the first product type set and the number of second reference product types in the second product type set, the first reference product type being the product type corresponding to the majority in the first product type set, and the second reference product type being the product type corresponding to the majority in the second product type set;

   Determining a first quantity ratio according to the total quantity of the first product type and the quantity of the first reference product type;

   determining a second quantity ratio according to the total quantity of the second product type and the quantity of the second reference product type;

   Determine the weight ratios of the selection interaction and the question-and-answer interaction, wherein the question-and-answer interaction has a greater weight than the selection interaction;

   Calculate the scores of the first reference product type and the second reference product type according to the first quantity proportion, the second quantity proportion, and the weight proportion;

   Determine the product type with a higher score between the first reference product type and the second reference product type as the target product type.
4. The method according to claim 3, characterized in that the step of determining the product type with a higher score among the first reference product type and the second reference product type as the target product type comprises:

   Determine the product type with the high score as the candidate product type;

   Obtaining historical recommendation records, where the historical recommendation records are recommendation records for the product corresponding to the target plot;

   Get the recommendation adoption status corresponding to each historical recommendation record;

   Determining whether the historical recommendation record includes the candidate product type;

   If yes, determining the recommendation adoption rate of the candidate product type according to the recommendation adoption situation;

   When the recommended adoption rate is higher than a preset adoption rate, the candidate product type is determined to be the target product type.
5. The method according to claim 4, characterized in that the method further comprises:

   In a case where the recommendation adoption rate is not higher than the preset adoption rate, determining whether the product type included in the most recent target interaction record is the same as the candidate product type;

   If they are the same, determining the product type with the highest recommendation adoption rate according to the recommendation adoption situation, and determining the candidate product type and the product type with the highest recommendation adoption rate as the target product type;

   If they are not the same, the candidate product type and the product type included in the most recent target interaction record are determined to be the target product type.
6. The method according to any one of claims 1 to 5, characterized in that generating recommendation information according to the product information comprises:

   Determining, from the target interaction records, an interaction record corresponding to the target product type as a final interaction record;

   Acquire final plot information included in the final interaction record, wherein the final plot information includes character information and time information;

   generating a first recommendation statement according to the character information, the time information and the target product type;

   generating a second recommendation statement according to the target product;

   generating a final recommendation statement according to the first recommendation statement and the second recommendation statement;

   The recommendation information is generated according to the final recommendation sentence and the product information.
7. The method according to claim 6, characterized in that the step of determining the target product according to the target product type comprises:

   Determining user information according to the target service type;

   Determine candidate products according to the target product type and the user information;

   Determining whether the target service type is suitable for repeated recommendations for the same product;

   If appropriate, determining the target product from the candidate products;

   If not suitable, the recommended products among the candidate products are deleted, and the target product is determined from the remaining candidate products.
8. A product recommendation device based on interactive novels, characterized in that it is applied to an electronic device in an integrated service system, wherein the integrated service system includes the electronic device and a server, the electronic device is communicatively connected with the server, and the electronic device is provided with a voice interactive service engine; the device includes:

   A first acquisition unit is used to call the voice interaction service engine to acquire the user's target voice information, perform intent recognition on the target voice information, and obtain the user's intent recognition result, wherein the intent recognition result is used to characterize the user's demand intention;

   A first determining unit is used to determine a target service type adapted to the user according to the intention recognition result, wherein the target service type is a service type of at least one of the following service types supported by the server of the integrated service system: a film and television recommendation service, a food recommendation service, and a scenic spot recommendation service;

   A second acquisition unit is used to acquire the interaction record of the user in the interactive novel, wherein the interaction record includes the interactive voice information input by the user and the plot content of the interactive novel corresponding to the interactive voice information;

   A second determining unit, configured to determine a target interaction record associated with the target service type from the interaction record according to the plot content;

   A third determining unit is used to determine a target product type according to the target interaction record, wherein the target product type is at least one product type among a plurality of product types corresponding to the target service type, and the product type is used to indicate a classification of products included in the service corresponding to the target service type;

   A fourth determining unit, configured to determine a target product according to the target product type, wherein the target product is a product belonging to the target product type;

   a third acquiring unit, configured to acquire product information about the target product from the server;

   An output unit is used to generate recommendation information based on the product information and call the voice interaction service engine to output the recommendation information.
9. An electronic device, characterized in that it includes a processor, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the processor, and the program includes instructions for executing the steps in the method according to any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program/instruction stored thereon, characterized in that the computer program/instruction is executed by a processor to implement the steps of the method according to any one of claims 1 to 7.