Electric power service recommendation method and device, equipment, storage medium and program product

Abstract

The invention relates to a power service recommendation method. The method comprises the following steps: acquiring first power consumption data of a target user and second power consumption data of a candidate user; converting a first feature of the first power consumption data into a first user tag, and converting a second feature of the second power consumption data into a second user tag; determining the similarity between the first user tag and the second user tag, obtaining the tag similarity corresponding to the candidate user, determining the similarity between the first power consumption data and the second power consumption data, obtaining the data similarity corresponding to the candidate user, and fusing the tag similarity and the data similarity to obtain the fusion similarity of the candidate user; based on the fusion similarity of the candidate users, selecting candidate users meeting a high similarity condition from the candidate users as neighbor users of the target user; and based on the satisfaction scores of the neighbor users on the candidate power services, selecting a target power service meeting a high-score condition from the candidate power services, and recommending the target power service to the target user. By adopting the method, the service recommendation accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technologies, and particularly to a power service recommendation method, apparatus, device, storage medium, and program product. Background Art

[0002] In power scenarios, it is often necessary to recommend power services that a power user has not used. Accurately recommending power services can not only improve user experience and satisfaction, but also increase service utilization rate, optimize resource allocation, enhance user loyalty, and promote innovation and development in the power market. Therefore, building a precise power service recommendation system is crucial for power companies.

[0003] In traditional technologies, power services with relatively high current popularity are usually randomly recommended to power users. However, power services with relatively high current popularity are sometimes not the services that users are truly interested in, and the recommendation accuracy rate of power services is relatively low. Summary of the Invention

[0004] Based on this, it is necessary to provide a power service recommendation method, apparatus, device, storage medium, and program product that can improve the recommendation accuracy rate of power services for the above technical problems.

[0005] In a first aspect, the present application provides a power service recommendation method, and the method includes:

[0006] Obtain first power consumption data of a target user and second power consumption data of at least one candidate user;

[0007] Convert a first feature of the first power consumption data into a first user label of the target user, and convert a second feature of the second power consumption data into a second user label of the candidate user;

[0008] For each candidate user, determine the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user, determine the similarity between the first power consumption data of the target user and the second power consumption data of the candidate user to obtain the data similarity corresponding to the candidate user, and fuse the label similarity and the data similarity corresponding to the candidate user to obtain the fusion similarity corresponding to the candidate user;

[0009] Based on the fusion similarity corresponding to each of the at least one candidate user, select a candidate user that meets the high similarity condition from the at least one candidate user as a neighbor user of the target user;

[0010] Based on the satisfaction scores of the preset at least one candidate power service by the neighbor users, select a target power service that meets the high score condition from the at least one candidate power service, and recommend the target power service to the target user.

[0011] In a second aspect, the present application provides a power service recommendation device, and the device includes:

[0012] An acquisition module, configured to acquire first power consumption data of a target user and second power consumption data of at least one candidate user;

[0013] A conversion module, configured to convert a first feature of the first power consumption data into a first user label of the target user, and convert a second feature of the second power consumption data into a second user label of the candidate user;

[0014] A determination module, configured to, for each candidate user, determine the similarity between the first user label of the target user and the second user label of the candidate user, obtain the label similarity corresponding to the candidate user, determine the similarity between the first power consumption data of the target user and the second power consumption data of the candidate user, obtain the data similarity corresponding to the candidate user, and fuse the label similarity and the data similarity corresponding to the candidate user to obtain the fusion similarity corresponding to the candidate user;

[0015] A selection module, configured to, based on the fusion similarity corresponding to each of the at least one candidate user, select a candidate user that meets the high similarity condition from the at least one candidate user as a neighbor user of the target user;

[0016] The selection module is further configured to, based on the satisfaction scores of the preset at least one candidate power service by the neighbor users, select a target power service that meets the high score condition from the at least one candidate power service, and recommend the target power service to the target user.

[0017] In a third aspect, the present application provides a computer device, including a memory and a processor, where a computer program is stored in the memory, and when the processor executes the computer program, the steps in the method embodiments of the present application are implemented.

[0018] In a fourth aspect, the present application provides a computer-readable storage medium, storing a computer program, and when the computer program is executed by a processor, the steps in the method embodiments of the present application are implemented.

[0019] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps in the method embodiments of the present application are implemented.

[0020] The above-mentioned power service recommendation method, device, equipment, storage medium, and program product obtain the first power consumption data of a target user and the second power consumption data of at least one candidate user; convert the first feature of the first power consumption data into the first user label of the target user, and convert the second feature of the second power consumption data into the second user label of the candidate user; for each candidate user, determine the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user, determine the similarity between the first power consumption data of the target user and the second power consumption data of the candidate user to obtain the data similarity corresponding to the candidate user, and fuse the label similarity and data similarity corresponding to the candidate user to obtain the fusion similarity corresponding to the candidate user; based on the fusion similarity corresponding to each of the at least one candidate user, select, from the at least one candidate user, the candidate user that meets the high similarity condition as the neighbor user of the target user; based on the satisfaction scores of the neighbor users for at least one preset candidate power service, select the target power service that meets the high score condition from the at least one candidate power service, and recommend the target power service to the target user. Compared with the traditional power service recommendation method, the present application constructs corresponding user labels for users based on the features of the users' power consumption data, fuses the label similarity of the user labels and the data similarity of the power consumption data, and based on the fused fusion similarity, screens out the power users with relatively high similarity to the target power user from many power users, and based on the satisfaction scores of the selected similar users for each power service, screens out the power services that the target user is likely to be interested in from each power service. In this way, personalized power service recommendation can be realized for specific users, recommend the power services that the users are really interested in to the users, and improve the accuracy of power service recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is an application environment diagram of the power service recommendation method in an embodiment;

[0022] Figure 2 It is a flowchart of the power service recommendation method in an embodiment;

[0023] Figure 3 It is a schematic diagram of the user label system in an embodiment;

[0024] Figure 4 It is a structural block diagram of the power service recommendation device in an embodiment;

[0025] Figure 5 It is an internal structure diagram of a computer device in an embodiment;

[0026] Figure 6 It is an internal structure diagram of a computer device in another embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In order to make the objectives, technical solutions, and advantages of this application more clear and understandable, the following further details this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not used to limit this application.

[0028] The power service recommendation method provided by this application can be applied to an application environment as Figure 1 shown. Among them, the terminal 102 communicates with the server 104 through a network. The data storage system can be set up separately and can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be placed on the cloud or other servers. Among them, the terminal 102 can be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, vehicle-mounted terminals, intelligent voice interaction devices, aircraft, smart home appliances, and portable wearable devices. The smart home appliances can be smart speakers, smart TVs, smart air conditioners, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or can also be a cloud server that provides network security services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, cloud security, host security, CDN, as well as basic cloud computing services such as big data and artificial intelligence platforms. The terminal 102 and the server 104 can be directly or indirectly connected through wired or wireless communication methods, and this application does not make any restrictions here.

[0029] The server 104 can obtain the first power consumption data of the target user and the second power consumption data of at least one candidate user, and convert the first feature of the first power consumption data into the first user label of the target user, and convert the second feature of the second power consumption data into the second user label of the candidate user. For each candidate user, the server 104 can determine the similarity between the first user label of the target user and the second user label of the candidate user, obtain the label similarity corresponding to the candidate user, determine the similarity between the first power consumption data of the target user and the second power consumption data of the candidate user, obtain the data similarity corresponding to the candidate user, and fuse the label similarity and the data similarity corresponding to the candidate user to obtain the fusion similarity corresponding to the candidate user. The server 104 can select, based on the fusion similarity corresponding to each of the at least one candidate user, the candidate user that meets the high similarity condition from the at least one candidate user as the neighbor user of the target user. The server 104 can select, based on the satisfaction scores of the neighbor users for at least one preset candidate power service, the target power service that meets the high score condition from the at least one candidate power service, and recommend the target power service to the terminal 102 held by the target user.

[0030] It can be understood that this embodiment does not limit this, and it can be understood that Figure 1 the application scenarios in

[0031] In one embodiment, as Figure 2 shown, a power service recommendation method is provided. This method can be applied to a computer device, and the computer device can be a terminal or a server. That is, this method can be executed independently by the terminal or the server, or can be implemented through the interaction between the terminal and the server. This embodiment takes the application of this method to a computer device as an example for illustration, including the following steps:

[0032] Step 202, obtain the first power consumption data of the target user and the second power consumption data of at least one candidate user.

[0033] In one embodiment, the computer device can collect the first power consumption data of the target user and the second power consumption data of at least one candidate user from multiple power service platforms (i.e., multiple data sources). Among them, the target user and the candidate user are both power consumption users in the power scenario.

[0034] In one embodiment, the multiple power service platforms may include a customer service platform, a power grid management platform, a customer center, a metering automation system, etc. The first power consumption data and the second power consumption data may include the user's profile data (such as age, gender, power consumption location, etc.), power consumption information, power consumption service handling information, payment information, metering device information, service evaluation information, etc.

[0035] Step 204, convert the first feature of the first power consumption data into the first user label of the target user, and convert the second feature of the second power consumption data into the second user label of the candidate user.

[0036] In one embodiment, the computer device can extract features from the first power consumption data to obtain the first feature of the first power consumption data, and extract features from the second power consumption data to obtain the second feature of the second power consumption data. Furthermore, the computer device can convert the first feature of the first power consumption data into the first user label of the target user according to a preset rule, and convert the second feature of the second power consumption data into the second user label of the candidate user. For example, by analyzing the monthly power consumption of users, the clustering algorithm is used to divide users into different consumption levels, and the users are labeled to obtain user labels.

[0037] In one embodiment, as Figure 3As shown in the figure, combined with the main services of power supply services, a user label system is constructed using a tree structure. Among them, the first-level labels in the user label system can be divided into basic, business expansion, meter reading and billing, customer service, and value-added, etc. The second, third, and fourth-level labels are their subordinate labels. For example, the second-level labels of the basic attribute labels can include power attributes and social attributes, the third-level labels can include document types, electricity consumption types, and nationalities, and the fourth-level labels can include identity cards and household registers.

[0038] Step 206: For each candidate user, determine the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user. Determine the similarity between the first electricity consumption data of the target user and the second electricity consumption data of the candidate user to obtain the data similarity corresponding to the candidate user. Integrate the label similarity and data similarity corresponding to the candidate user to obtain the integrated similarity corresponding to the candidate user.

[0039] In one embodiment, for each candidate user, the computer device can calculate the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user. The computer device can calculate the similarity between the first electricity consumption data of the target user and the second electricity consumption data of the candidate user to obtain the data similarity corresponding to the candidate user. Furthermore, the computer device can integrate the label similarity and data similarity corresponding to the candidate user to obtain the integrated similarity corresponding to the candidate user.

[0040] In one embodiment, determining the similarity between the first electricity consumption data of the target user and the second electricity consumption data of the candidate user to obtain the data similarity corresponding to the candidate user includes: extracting features from the first electricity consumption data of the target user to obtain the data features of the target user; for each candidate user, extracting features from the second electricity consumption data of the candidate user to obtain the data features of the candidate user; determining the similarity between the data features of the target user and the data features of the candidate user to obtain the data similarity corresponding to the candidate user.

[0041] In one embodiment, the integrated similarity can be calculated through the following formula:

[0042] ;

[0043] Among them, represents the label similarity between the target user u and the candidate user v, represents the data similarity between the target user u and the candidate user v, a represents the integration weight of the label similarity, (1 - a) represents the integration weight of the data similarity, represents the integrated similarity.

[0044] Step 208: Based on the respective fusion similarities of at least one candidate user, select, from the at least one candidate user, the candidate users that meet the high similarity condition as the neighbor users of the target user.

[0045] In one embodiment, the computer device may select, from the at least one candidate user, the candidate users with a fusion similarity greater than a preset similarity threshold as the neighbor users of the target user.

[0046] Step 210: Based on the satisfaction scores of the neighbor users for at least one preset candidate power service, select, from the at least one candidate power service, the target power service that meets the high score condition, and recommend the target power service to the target user.

[0047] In one embodiment, for each candidate power service in the at least one preset candidate power service, the computer device may use the respective fusion similarity of each neighbor user as the scoring weight for the satisfaction score of the neighbor user for this candidate power service, so as to perform a weighted average on the satisfaction scores of each neighbor user for this candidate power service to obtain the comprehensive score corresponding to this candidate power service. Furthermore, the computer device may select, from the at least one candidate power service, the candidate power service with a comprehensive score greater than a preset score threshold as the target power service, and recommend the selected target power service to the target user.

[0048] In the above power service recommendation method, the first power consumption data of the target user and the second power consumption data of at least one candidate user are obtained; the first feature of the first power consumption data is converted into the first user label of the target user, and the second feature of the second power consumption data is converted into the second user label of the candidate user; for each candidate user, the similarity between the first user label of the target user and the second user label of the candidate user is determined to obtain the label similarity corresponding to the candidate user, the similarity between the first power consumption data of the target user and the second power consumption data of the candidate user is determined to obtain the data similarity corresponding to the candidate user, and the label similarity and the data similarity corresponding to the candidate user are fused to obtain the fusion similarity corresponding to the candidate user; based on the fusion similarities corresponding to at least one candidate user respectively, from at least one candidate user, the candidate users meeting the high similarity condition are selected as the neighbor users of the target user; based on the satisfaction scores of the neighbor users for at least one preset candidate power service, the target power service meeting the high score condition is selected from at least one candidate power service, and the target power service is recommended to the target user. Compared with the traditional power service recommendation method, in this application, corresponding user labels are constructed for users based on the features of the power consumption data of the users, the label similarity of the user labels and the data similarity of the power consumption data are fused, and based on the fused fusion similarity, the power consumption users with higher similarity to the target power consumption user are screened out from many power consumption users, and based on the satisfaction scores of the screened similar users for each power service, the power services that the target user is likely to be interested in are screened out from each power service. In this way, personalized power service recommendation can be realized for specific users, the power services that the users are really interested in can be recommended to the users, and the accuracy of power service recommendation can be improved.

[0049] In one embodiment, obtaining the first power consumption data of the target user and the second power consumption data of at least one candidate user includes: collecting the first original power consumption data of the target user and the second original power consumption data of at least one candidate user from multiple power service platforms; the first original power consumption data includes a target user identifier for uniquely identifying the target user, and the second original power consumption data includes a candidate user identifier for uniquely identifying the candidate user; based on the target user identifier of the target user, the first original power consumption data from multiple power service platforms is summarized to obtain the first summarized power consumption data of the target user, and the first summarized power consumption data is cleaned to obtain the first power consumption data of the target user; for each candidate user, based on the candidate user identifier of the candidate user, the second original power consumption data from multiple power service platforms is summarized to obtain the second summarized power consumption data of the candidate user, and the second summarized power consumption data is cleaned to obtain the second power consumption data of the candidate user.

[0050] In one embodiment, the user identifier may specifically be the user's power consumption account number or mobile phone number.

[0051] In one embodiment, the aggregated power consumption data is cleaned, which may specifically include duplicate data removal processing and abnormal data deletion processing. Among them, the duplicate data removal processing can specifically determine the duplication situation based on the key identification fields in the data. For example, through comprehensive judgment by combining the user identification with other relevant fields such as the timestamp and operation type, only one copy of the duplicate record data is retained, and the rest are deleted, so as to ensure the uniqueness of the data and avoid interference from duplicate data on subsequent service recommendation analysis. The abnormal data deletion processing can specifically use visualization tools such as box plots to assist in observing the data distribution, intuitively find the interval where the abnormal values in the data are located, and then perform the elimination operation to ensure that the data used for calculating user tags conforms to the normal business logic and user behavior patterns as much as possible.

[0052] In the above embodiment, by collecting the power consumption data of users from multiple power service platforms and aggregating the power consumption data from multiple power service platforms according to the user identification that uniquely identifies the user, the richness of the user power consumption data can be improved. By cleaning the aggregated power consumption data, redundant data and abnormal data can be filtered out, and the influence of redundant data or abnormal data on the subsequent service recommendation analysis process can be avoided, further improving the accuracy of power service recommendations.

[0053] In one embodiment, determining the similarity between the first user tag of the target user and the second user tag of the candidate user to obtain the tag similarity corresponding to the candidate user includes: converting the first user tag of the target user into a vector form to obtain the first tag vector of the target user; converting the second user tag of the candidate user into a vector form to obtain the second tag vector of the candidate user; determining the similarity between the first tag vector of the target user and the second tag vector of the candidate user to obtain the tag similarity corresponding to the candidate user.

[0054] For example, if each user can have 10 different user tags, for a certain user, if the user has three user tags of "low-voltage user", "power outage once this month", and "complaint-sensitive", then its tag vector can be expressed as [0, 1, 0, 1, 0, 0, 0, 1, 0, 0], where 1 indicates that the user has the user tag, and 0 indicates that the user does not have the user tag.

[0055] In one embodiment, the computer device can use cosine similarity to represent the tag similarity, and the calculation formula of cosine similarity is as follows:

[0056] ;

[0057] where u and v are two users, and are the weights of user u and user v for tag i respectively, and n is the total number of user tags. represents the cosine similarity.

[0058] In the above embodiments, by converting user tags into tag vectors that are convenient for calculation, and obtaining the tag similarity corresponding to the candidate user by calculating the similarity between the tag vector of the target user and the tag vector of the candidate user, the accuracy of the tag similarity can be improved.

[0059] In one embodiment, based on the respective fusion similarities corresponding to at least one candidate user, selecting, from the at least one candidate user, the candidate user that meets the high similarity condition as the neighbor user of the target user includes: sorting the at least one candidate user in descending order according to the respective fusion similarities corresponding to the at least one candidate user to obtain a sorted candidate user sequence; selecting the first preset number of candidate users from the candidate user sequence as the neighbor users of the target user.

[0060] In the above embodiments, by selecting the first preset number of candidate users from the candidate user sequence sorted in descending order of fusion similarity as the neighbor users of the target user, the neighbor users most similar to the target user can be selected, thereby further improving the accuracy of subsequent service recommendation based on the satisfaction scores of the neighbor users for the candidate power services.

[0061] In one embodiment, based on the satisfaction scores of the neighbor users for at least one preset candidate power service, selecting the target power service that meets the high score condition from the at least one candidate power service includes: for each candidate power service in the at least one preset candidate power service, using the respective fusion similarities corresponding to the neighbor users as the scoring weights for the satisfaction scores of the neighbor users for the candidate power service, so as to perform a weighted average on the satisfaction scores of the neighbor users for the candidate power service to obtain the comprehensive score corresponding to the candidate power service; sorting the at least one candidate power service in descending order according to the respective comprehensive scores corresponding to the at least one candidate power service to obtain a sorted candidate power service sequence; selecting the first preset number of candidate power services from the candidate power service sequence as the target power service that meets the high score condition.

[0062] In one embodiment, the comprehensive score can be calculated by the following formula:

[0063] ;

[0064] where, N(u) represents the set of neighbor users v composed of the neighbor users of the target user u, represents the fusion similarity between the neighbor user v and the target user u, is the satisfaction score of neighbor user v for power service i, is the comprehensive score for power service i.

[0065] In the above embodiments, by using the fusion similarity of neighbor users as the scoring weight for the satisfaction score of neighbor users for candidate power services to perform weighted averaging on the satisfaction scores, higher weights can be assigned to more similar neighbor users, thereby improving the accuracy of the comprehensive score. And by selecting the candidate power services with the top preset number of power services from the candidate power service sequence sorted from high to low in terms of the comprehensive score as the target power services, the target power services that the target user may be most interested in can be selected, further improving the accuracy of service recommendation.

[0066] In one embodiment, the method further includes: obtaining feedback data of the target user for the recommended target power service, and optimizing the first user label of the target user based on the feedback data.

[0067] In one embodiment, in addition to being able to optimize the first user label of the target user based on the feedback data, the computer device can also optimize parameters such as the similarity calculation weight, the number of neighbor users, and the number of recommended services based on the feedback data to improve the service recommendation effect.

[0068] In the above embodiments, by optimizing the first user label of the target user based on the feedback data, the target user can have a more accurate user label, thereby further improving the accuracy of subsequent power service recommendation based on its user label.

[0069] It should be understood that although the steps in the flowcharts of the above embodiments are shown in sequence one by one, these steps are not necessarily executed in sequence. Unless there is a clear indication in this article, the execution of these steps has no strict sequence limit, and these steps can be executed in other sequences. Moreover, at least a part of the steps in the above embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential either, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0070] In one embodiment, as Figure 4 shown, a power service recommendation device 400 is provided, and the device specifically includes:

[0071] An acquisition module 402, configured to acquire first power consumption data of a target user and second power consumption data of at least one candidate user;

[0072] A conversion module 404 is configured to convert the first feature of the first electricity consumption data into a first user label of the target user, and convert the second feature of the second electricity consumption data into a second user label of the candidate user.

[0073] A determination module 406 is configured to, for each candidate user, determine the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user, determine the similarity between the first electricity consumption data of the target user and the second electricity consumption data of the candidate user to obtain the data similarity corresponding to the candidate user, and fuse the label similarity and the data similarity corresponding to the candidate user to obtain the fused similarity corresponding to the candidate user.

[0074] A selection module 408 is configured to, based on the fused similarity corresponding to each of at least one candidate user, select, from the at least one candidate user, the candidate user that meets the high-similarity condition as the neighbor user of the target user.

[0075] The selection module 408 is further configured to, based on the satisfaction scores of the neighbor users for at least one candidate power service preset, select the target power service that meets the high-score condition from the at least one candidate power service, and recommend the target power service to the target user.

[0076] In one embodiment, the acquisition module 402 is further configured to collect the first original electricity consumption data of the target user and the second original electricity consumption data of at least one candidate user from multiple power service platforms; the first original electricity consumption data includes a target user identifier for uniquely identifying the target user, and the second original electricity consumption data includes a candidate user identifier for uniquely identifying the candidate user; based on the target user identifier of the target user, summarize the first original electricity consumption data from multiple power service platforms to obtain the first summarized electricity consumption data of the target user, and clean the first summarized electricity consumption data to obtain the first electricity consumption data of the target user; for each candidate user, based on the candidate user identifier of the candidate user, summarize the second original electricity consumption data from multiple power service platforms to obtain the second summarized electricity consumption data of the candidate user, and clean the second summarized electricity consumption data to obtain the second electricity consumption data of the candidate user.

[0077] In one embodiment, the determination module 406 is further configured to convert the first user label of the target user into a vector form to obtain the first label vector of the target user; convert the second user label of the candidate user into a vector form to obtain the second label vector of the candidate user; and determine the similarity between the first label vector of the target user and the second label vector of the candidate user to obtain the label similarity corresponding to the candidate user.

[0078] In one embodiment, the selection module 408 is further configured to sort at least one candidate user in descending order according to the respective fusion similarity corresponding to each of the at least one candidate user, so as to obtain a sorted candidate user sequence; and select the candidate users with the top preset number of users from the candidate user sequence as the neighbor users of the target user.

[0079] In one embodiment, the selection module 408 is further configured to, for each of the preset at least one candidate power service, use the respective fusion similarity corresponding to each neighbor user as the scoring weight for the satisfaction score of each neighbor user for the candidate power service, so as to perform a weighted average on the satisfaction scores of each neighbor user for the candidate power service to obtain a comprehensive score corresponding to the candidate power service; sort the at least one candidate power service in descending order according to the respective comprehensive score corresponding to each of the at least one candidate power service, so as to obtain a sorted candidate power service sequence; and select the candidate power services with the top preset number of power services from the candidate power service sequence as the target power services that meet the high-score condition.

[0080] In one embodiment, the apparatus further includes:

[0081] An optimization module, configured to obtain feedback data of the target user on the recommended target power service, and optimize the first user label of the target user based on the feedback data.

[0082] The above-mentioned power service recommendation device obtains the first electricity consumption data of a target user and the second electricity consumption data of at least one candidate user; converts the first feature of the first electricity consumption data into a first user label of the target user, and converts the second feature of the second electricity consumption data into a second user label of the candidate user; for each candidate user, determines the similarity between the first user label of the target user and the second user label of the candidate user, obtains the label similarity corresponding to the candidate user, determines the similarity between the first electricity consumption data of the target user and the second electricity consumption data of the candidate user, obtains the data similarity corresponding to the candidate user, and fuses the label similarity and the data similarity corresponding to the candidate user to obtain the fusion similarity corresponding to the candidate user; based on the fusion similarity corresponding to each of the at least one candidate user, selects, from the at least one candidate user, the candidate user that meets the high similarity condition as the neighbor user of the target user; based on the satisfaction scores of the neighbor users for at least one preset candidate power service, selects, from the at least one candidate power service, the target power service that meets the high score condition, and recommends the target power service to the target user. Compared with the traditional power service recommendation method, in this application, corresponding user labels are constructed for users based on the characteristics of the users' electricity consumption data, the label similarity of the user labels and the data similarity of the electricity consumption data are fused, and based on the fused similarity, electricity users with relatively high similarity to the target electricity user are screened out from numerous electricity users, and based on the satisfaction scores of the screened similar users for each power service, the power services that the target user is likely to be interested in are screened out from each power service. In this way, personalized power service recommendation can be realized for specific users, the power services that the users are really interested in can be recommended to the users, and the accuracy of power service recommendation can be improved.

[0083] Each module in the above-mentioned power service recommendation device can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0084] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 5As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it realizes a power service recommendation method.

[0085] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 6 shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it realizes a power service recommendation method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad set on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0086] Those skilled in the art can understand that Figure 5 and Figure 6The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component layout.

[0087] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0088] In one embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0089] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0090] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions.

[0091] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the above method embodiments. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application may include at least one of non-volatile and volatile memories. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0092] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0093] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A power service recommendation method, characterized in that, The method includes: Obtaining first electricity consumption data of a target user and second electricity consumption data of at least one candidate user; Converting a first feature of the first electricity consumption data into a first user label of the target user, and converting a second feature of the second electricity consumption data into a second user label of the candidate user; For each candidate user, determining the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user, determining the similarity between the first electricity consumption data of the target user and the second electricity consumption data of the candidate user to obtain the data similarity corresponding to the candidate user, and fusing the label similarity and the data similarity corresponding to the candidate user to obtain the fusion similarity corresponding to the candidate user; Based on the fusion similarities respectively corresponding to the at least one candidate user, selecting candidate users that meet the high similarity condition from the at least one candidate user as the neighbor users of the target user; Based on the satisfaction scores of the neighbor users for at least one candidate power service, selecting a target power service that meets the high score condition from the at least one candidate power service, and recommending the target power service to the target user.

2. The method according to claim 1, wherein The obtaining of the first electricity consumption data of the target user and the second electricity consumption data of at least one candidate user includes: Collecting first original electricity consumption data of the target user and second original electricity consumption data of at least one candidate user from multiple power service platforms; the first original electricity consumption data contains a target user identifier for uniquely identifying the target user, and the second original electricity consumption data contains a candidate user identifier for uniquely identifying the candidate user; Based on the target user identifier of the target user, summarizing the first original electricity consumption data from multiple power service platforms to obtain the first summarized electricity consumption data of the target user, and cleaning the first summarized electricity consumption data to obtain the first electricity consumption data of the target user; For each candidate user, based on the candidate user identifier of the candidate user, summarizing the second original electricity consumption data from multiple power service platforms to obtain the second summarized electricity consumption data of the candidate user, and cleaning the second summarized electricity consumption data to obtain the second electricity consumption data of the candidate user.

3. The method according to claim 1, characterized in that, The determining of the similarity between the first user label of the target user and the second user label of the candidate user to obtain the label similarity corresponding to the candidate user includes: Converting the first user label of the target user into a vector form to obtain the first label vector of the target user; Converting the second user label of the candidate user into a vector form to obtain the second label vector of the candidate user; Determining the similarity between the first label vector of the target user and the second label vector of the candidate user to obtain the label similarity corresponding to the candidate user.

4. The method according to claim 1, characterized in that, The selecting of candidate users that meet the high similarity condition from the at least one candidate user as the neighbor users of the target user based on the fusion similarities respectively corresponding to the at least one candidate user includes: Sort the at least one candidate user in descending order according to the respective corresponding fusion similarity, to obtain a sorted candidate user sequence; From the candidate user sequence, select the candidate users with the top preset number of users as the neighbor users of the target user.

5. The method according to claim 1, wherein The selecting, from the at least one candidate power service, a target power service that meets the high-score condition based on the satisfaction scores of the neighbor users for the preset at least one candidate power service includes: For each candidate power service in the preset at least one candidate power service, use the respective corresponding fusion similarity of each neighbor user as the scoring weight for the satisfaction score of the neighbor user for the candidate power service, so as to perform weighted averaging on the satisfaction scores of each neighbor user for the candidate power service, and obtain the comprehensive score corresponding to the candidate power service; Sort the at least one candidate power service in descending order according to the respective corresponding comprehensive score, to obtain a sorted candidate power service sequence; From the candidate power service sequence, select the candidate power services with the top preset number of power services as the target power services that meet the high-score condition.

6. The method according to any one of claims 1 to 5, characterized in that The method further includes: Obtain the feedback data of the target user for the recommended target power service, and optimize the first user label of the target user based on the feedback data.

7. An electric power service recommendation device, characterized in that The device includes: An acquisition module, configured to acquire the first power consumption data of the target user and the second power consumption data of at least one candidate user; A conversion module, configured to convert the first feature of the first power consumption data into the first user label of the target user, and convert the second feature of the second power consumption data into the second user label of the candidate user; A determination module, configured to, for each candidate user, determine the similarity between the first user label of the target user and the second user label of the candidate user, to obtain the label similarity corresponding to the candidate user, determine the similarity between the first power consumption data of the target user and the second power consumption data of the candidate user, to obtain the data similarity corresponding to the candidate user, and fuse the label similarity and the data similarity corresponding to the candidate user, to obtain the fusion similarity corresponding to the candidate user; A selection module, configured to, based on the respective corresponding fusion similarity of the at least one candidate user, select the candidate users that meet the high-similarity condition from the at least one candidate user as the neighbor users of the target user; The selection module is further configured to, based on the satisfaction scores of the neighbor users for the preset at least one candidate power service, select a target power service that meets the high-score condition from the at least one candidate power service, and recommend the target power service to the target user.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images