Personal portrait-based lease contract intelligent matching recommendation method and device

Abstract

The invention relates to the technical field of data processing, in particular to a personal portrait-based lease contract intelligent matching recommendation method and device. The recommendation method comprises the following steps: acquiring subjective information and objective information of a target user; the subjective information is information collected through active behaviors of the target user, and the objective information is information collected through passive behaviors of the target user; establishing a user portrait of the target user based on the subjective information and the objective information of the target user; obtaining feature description of the rental item based on the attribute of the rental item; calculating and obtaining a similarity score between the user portrait of the target user and the feature description of the rental item; and selecting the lease item with the highest similarity as a recommendation result according to the similarity score. The technical problems of insufficient personalization, inaccurate matching and insufficient risk control in an existing lease recommendation system are mainly solved by comprehensively applying a user portrait technology and lease item attribute matching.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for intelligently matching and recommending lease contracts based on personal portraits. Background Art

[0002] In the traditional rental market, users often face the problems of information overload and inaccurate matching. There are many types of rental items on the market, and users need to spend a lot of time and energy to find the most suitable option for themselves. Although existing rental platforms provide a large amount of housing or equipment information, the recommendation system is mostly based on simple keyword matching or popular sorting, which fails to fully consider the personalized needs and preferences of users, resulting in inaccurate recommendation results. In addition, traditional recommendation systems rely more on objective data (such as browsing history, transaction records), while ignoring the subjective wishes of users (such as living preferences, living habits, etc.), which may lead to recommendation results that do not match the actual needs of users. For some high-value rental items (such as mobile phones and cars), there is a lack of effective credit assessment mechanism, which increases the risk of lessors.

[0003] In recent years, with the development of big data and machine learning technology, user portrait technology has gradually matured and been applied in many fields. User portraits collect and analyze user behavior data and personal information to build a comprehensive and accurate user model, providing a solid foundation for personalized recommendations. However, the existing user portrait technology still has limitations when applied to the rental market: most systems only focus on the user's historical behavior data, and fail to comprehensively consider the user's subjective wishes and other potential factors; user portraits are static and cannot be dynamically adjusted to reflect the user's latest behavior changes; different types of rental objects have different attribute requirements, and the existing recommendation system fails to perform targeted optimization based on the specific characteristics of the rental objects.

[0004] In addition, existing rental recommendation systems also have shortcomings when dealing with rental contracts. Rental contracts involve multiple terms and conditions, and existing systems often fail to fully consider the impact of these terms on user choices, resulting in a gap between recommendation results and users' actual needs and expectations. At the same time, the lack of a prediction and management mechanism for possible problems that may arise during the long-term use of leased items further reduces the practicality and reliability of the recommendation system. Overall, the current rental recommendation system still has much room for improvement in personalization, accuracy, and risk control. Summary of the invention

[0005] This application mainly solves the technical problems of insufficient personalization, inaccurate matching and lack of risk control in existing rental recommendation systems by comprehensively applying user profiling technology and rental property matching.

[0006] An embodiment of the present application provides a rental contract intelligent matching recommendation method based on personal portraits, and the recommendation method includes: Collecting and obtaining subjective information and objective information of a target user; the subjective information is information collected through the active behavior of the target user, and the objective information is information collected through the passive behavior of the target user; Based on the subjective information and objective information of the target user, establish a user profile of the target user; Based on the attributes of the leased object, obtaining a characteristic description of the leased object; Calculate the similarity score between the user portrait of the target user and the feature description of the leased item; Based on the similarity score, the rental item with the highest similarity is selected as the recommendation result.

[0007] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the acquisition of subjective information and objective information of the target user includes: Collecting the subjective information of the target user through the active behavior of the target user; the initiative includes but is not limited to: questionnaire survey and / or actively filling in information; Objective information of the target user is collected through the passive behavior of the target user; the passive behavior includes but is not limited to: behavior data, and / or transaction records, and / or location service data, and / or credit records.

[0008] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the user portrait of the target user is established based on the subjective information and objective information of the target user, including: Based on the subjective information of the target user, a subjective information vector group of the target user is obtained; Based on the objective information of the target user, obtaining an objective information vector group of the target user; Based on the subjective information vector group and the objective information vector group, a user profile of the target user is constructed. .

[0009] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the subjective information vector group includes at least one dimensional subjective information vector, and the objective information vector group includes at least one dimensional objective information vector; The dimensions of the subjective information vector group and the objective information vector group are consistent.

[0010] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the user portrait of the target user It is expressed by the following formula: ; in, The target user User portraits, The target user The subjective information vector group, The target user Objective information vector group.

[0011] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the step of obtaining a characteristic description of the leased object based on the attributes of the leased object includes: The characteristic description of the leased property is expressed by the following formula : ; Subjective information vector group ; Objective information vector group ;

[0012] in, represents the k-dimensional feature vector, represents the k-dimensional subjective information vector, Represents the k-dimensional objective information vector.

[0013] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the calculation of the similarity score between the user portrait of the target user and the feature description of the leased item includes: Based on the properties of the leased object, different subjective information weights and objective information weights are configured for different leased objects; Obtaining a specific portrait of the target user; the specific portrait is a weighted combination of the subjective information weight and the objective information weight of the user portrait; Calculating a similarity score between a specific profile of the target user and a feature description of the rental item; Wherein, calculating the similarity score includes calculating cosine similarity or Euclidean distance.

[0014] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the specific portrait is represented by the following formula: ; in, Specific portraits of target users, is the subjective information weight, is the objective information weight.

[0015] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, when calculating the similarity score including calculating the cosine similarity, the cosine similarity is calculated by the following formula: ; When calculating the similarity score and the Euclidean distance, the Euclidean distance is calculated by the following formula: .

[0016] The present application also provides a rental contract intelligent matching recommendation device based on personal portraits, the intelligent matching recommendation device comprising: A collection module, the collection module is used to collect subjective information and objective information of the target user; the subjective information is information collected through the active behavior of the target user, and the objective information is information collected through the passive behavior of the target user; A user portrait building module, the user portrait building module is used to build a user portrait of the target user based on the subjective information and objective information of the target user; A feature description acquisition module, the feature description acquisition module is used to acquire a feature description of the leased object based on the attributes of the leased object; A similarity score calculation module, the similarity score calculation module is used to calculate a similarity score between a user portrait of a target user and a feature description of the leased item; The recommendation module is used to select the rental item with the highest similarity as the recommendation result according to the similarity score.

[0017] According to the above-mentioned embodiment of the intelligent matching recommendation method and device for leasing contracts based on personal portraits, a dynamic and comprehensive user portrait is constructed by comprehensively collecting the subjective information (such as questionnaires, social media analysis) and objective information (such as behavioral data, transaction records) of the target user to ensure its accuracy and timeliness, thereby solving the problems of information overload and inaccurate matching in the prior art; by obtaining feature descriptions based on the specific attributes of the leased object and calculating the similarity score between the user portrait and the leased object features, the recommendation system can accurately match the leased object that best meets the user's needs, significantly improving the accuracy and relevance of the recommendation and reducing the time cost of user screening and comparison; providing more personalized recommendation results, improving user experience and satisfaction, and solving the problem that traditional recommendation systems fail to fully consider the user's subjective wishes and other potential factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Flow chart (I) of the method for automatically transporting assembly parts provided in an embodiment of the present application; Figure 2Flow chart (II) of the automatic handling method for assembly parts provided in the embodiment of the present application; Figure 3 Flow chart of the automatic handling method of assembly parts provided in the embodiment of the present application (III); Figure 4 This is a functional block diagram of the automatic assembly handling device provided in an embodiment of the present application.

[0019] In the figure: 401, acquisition module; 402, user portrait establishment module; 403, feature description acquisition module; 404, similarity score calculation module; 405, recommendation module. DETAILED DESCRIPTION

[0020] The present application is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are intended to enable the present application to be better understood. However, those skilled in the art can easily recognize that some of the features can be omitted in different situations, or can be replaced by other elements, materials, and methods. In some cases, some operations related to the present application are not shown or described in the specification, in order to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations based on the description in the specification and the general technical knowledge in the art.

[0021] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0022] The serial numbers assigned to the components herein, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings).

[0023] Please refer to Figure 1 In order to solve the technical problems of insufficient personalization, inaccurate matching and lack of risk control in the existing rental recommendation system mainly by comprehensively applying user portrait technology and rental property matching, an embodiment of the present application provides a rental contract intelligent matching recommendation method based on personal portrait, and the recommendation method includes the following steps: Step S101: collect and obtain subjective information and objective information of a target user. The subjective information is information collected through the active behavior of the target user, and the objective information is information collected through the passive behavior of the target user.

[0024] In some embodiments, the subjective and objective information of the target user is comprehensively collected through multiple channels; subjective information is information collected through the user's active behavior, such as obtaining residential preferences, living habits, and hobbies through questionnaires, social media analysis, user feedback, etc.; objective information is data collected through the user's passive behavior, such as browsing history, search keywords, click behavior, transaction records, location service data, etc. on the platform. All data collection activities should strictly comply with relevant laws and regulations and obtain the user's explicit consent to ensure privacy protection.

[0025] Step S102: Create a user profile of the target user based on the subjective information and objective information of the target user.

[0026] In some embodiments, a dynamic and comprehensive user portrait is constructed using the subjective information and objective information collected in step S101; useful features such as geographic location preferences, budget range, life needs, etc. are extracted through feature engineering, and necessary preprocessing (such as standardization and normalization) is performed; the user portrait can be updated regularly to reflect the user's latest behavior and preference changes to maintain its timeliness and accuracy.

[0027] In some embodiments, a clustering algorithm may be used to divide users into different groups and assign meaningful labels to each group for subsequent personalized recommendations and services.

[0028] Step S103: Based on the attributes of the leased object, obtain a characteristic description of the leased object.

[0029] In some embodiments, specific feature descriptions are defined for different types of rental items (such as houses, mobile phones, and cars), such as geographic location, house type, rent, area, number of rooms, brand, model, performance parameters, etc.; the feature descriptions can be obtained from the rental platform, third-party databases, or suppliers, and the accuracy and completeness of the data are ensured.

[0030] In some embodiments, for categorical features (such as house type, brand), one-hot encoding or other encoding methods are used to convert them into numerical form to facilitate mathematical operations.

[0031] Step S104: Calculate a similarity score between the user portrait of the target user and the feature description of the leased item.

[0032] In some embodiments, common methods such as cosine similarity and Euclidean distance are used to calculate the similarity score between the user portrait and the feature description of the leased item. According to the attributes of different leased items, the weights of subjective information and objective information are dynamically adjusted to ensure that the recommendation results are in line with the user's subjective wishes and take into account objective conditions.

[0033] In some embodiments, a machine learning model (such as a random forest, a neural network) can be used to train a recommendation model to improve the accuracy and generalization ability of similarity calculations, thereby achieving more accurate matching.

[0034] Step S105: According to the similarity score, select the rental item with the highest similarity as the recommendation result.

[0035] In some embodiments, the candidate rental items are sorted according to the similarity scores, and the options that best meet the user's needs are presented to the user; in addition to the options that best meet the user's current preferences, some rental items of potential interest are also provided to discover the user's new needs.

[0036] In some embodiments, an instant feedback mechanism (such as "like" and "dislike" buttons) and detailed feedback channels (such as questionnaires or comment boxes) can also be set up to allow users to quickly provide feedback on recommendation results, which can be used to further optimize user portraits and recommendation methods.

[0037] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the collection of subjective information and objective information of the target user includes: collecting the subjective information of the target user through the active behavior of the target user; the initiative includes but is not limited to: questionnaire surveys and / or active filling in of information; collecting the objective information of the target user through the passive behavior of the target user; the passive behavior includes but is not limited to: behavioral data, and / or transaction records, and / or location service data, and / or credit records.

[0038] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the step S102 of establishing a user portrait of the target user based on the subjective information and objective information of the target user specifically includes the following steps: Step S201: Obtain a subjective information vector group of the target user based on the subjective information of the target user.

[0039] In some embodiments, by analyzing the subjective information of the target user, a subjective information vector group is extracted and constructed; this subjective information is data collected through the user's active behavior, such as residential preferences (such as geographical location, room size, decoration style), living habits (such as whether to keep pets, whether to smoke), and hobbies obtained through channels such as questionnaires, social media analysis, and user feedback. Finally, this information is converted into a structured feature vector to ensure that each dimension can accurately reflect the user's personalized needs and preferences.

[0040] Step S202: Based on the objective information of the target user, obtain an objective information vector group of the target user.

[0041] In some embodiments, an objective information vector group is constructed based on the objective information of the target user; this objective information is data collected through the user's passive behavior, such as browsing history, search keywords, click behavior, transaction records, location service data, etc. on the platform. Finally, these data are processed and analyzed to extract useful features and convert them into numerical or categorical feature vectors to fully describe the user's behavior patterns and actual needs.

[0042] Step S203: construct a user profile of the target user based on the subjective information vector group and the objective information vector group .

[0043] In some embodiments, the user profile of the target user is comprehensively constructed by combining the subjective information vector group and the objective information vector group. Through feature engineering, the feature vectors of subjective information and objective information are integrated to form a comprehensive and dynamic user model. This model not only reflects the user's explicit preferences (such as questionnaire survey results), but also covers implicit behavioral characteristics (such as browsing and clicking behaviors).

[0044] In the above-mentioned intelligent matching and recommendation method for leasing contracts based on personal portraits, as a preferred scheme, the subjective information vector group contains a subjective information vector of at least one dimension, and the objective information vector group contains an objective information vector of at least one dimension; the dimensions of the subjective information vector group and the objective information vector group are consistent.

[0045] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the user portrait of the target user It is expressed by the following formula: ; in, The target user User portraits, The target user The subjective information vector group, The target user Objective information vector group.

[0046] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the step of obtaining a characteristic description of the leased object based on the attributes of the leased object includes: The characteristic description of the leased property is expressed by the following formula : ; Subjective information vector group ; Objective information vector group ;

[0047] in, represents the k-dimensional feature vector, represents the k-dimensional subjective information vector, Represents the k-dimensional objective information vector.

[0048] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the calculation in step S104 to obtain the similarity score between the user portrait of the target user and the feature description of the leased item specifically includes the following steps: Step S301: Based on the properties of the leased object, different subjective information weights and objective information weights are configured for different leased objects.

[0049] In some embodiments, different subjective information weights and objective information weights are configured for different types of leased items according to their specific attributes. For example, for a house, subjective information such as geographic location and living preferences may be given more weight, so a higher subjective information weight is assigned; while for a mobile phone, objective information such as performance parameters and credit scores may be given more attention, so a higher objective information weight is assigned. This differentiated configuration ensures that the similarity calculation is more in line with the actual needs of different types of leased items.

[0050] Step S302: Obtain a specific portrait of the target user; the specific portrait is a weighted combination of the subjective information weight and the objective information weight of the user portrait.

[0051] In some embodiments, a specific profile of the target user is generated by weighted combination of the user's subjective information vector group and objective information vector group according to pre-configured weights. The specific profile comprehensively reflects the user's personalized needs and behavior patterns under the current rental type. The weighted combination process ensures that the final user profile not only contains the user's explicit preferences, but also covers implicit behavioral characteristics, thereby improving the relevance and accuracy of the recommendation results.

[0052] Step S303: Calculate the similarity score between the specific portrait of the target user and the feature description of the leased item, wherein calculating the similarity score includes calculating cosine similarity or Euclidean distance.

[0053] In some embodiments, a common method such as cosine similarity or Euclidean distance is used to calculate the similarity score between the specific portrait of the target user and the feature description of the rental item. Cosine similarity is used to measure the angle between two vectors and is suitable for processing high-dimensional sparse data; while Euclidean distance measures the straight-line distance between two points and is suitable for processing low-dimensional dense data. Selecting an appropriate similarity measurement method can ensure that the calculation of the similarity score is more accurate and reliable, thereby achieving precise matching. Finally, the candidate rental items are sorted according to the calculated similarity scores to determine the recommendation results that best meet the user's needs.

[0054] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, the specific portrait is represented by the following formula: ; in, Specific portraits of target users, is the subjective information weight, is the objective information weight.

[0055] In the above-mentioned intelligent matching recommendation method for leasing contracts based on personal portraits, as a preferred solution, when calculating the similarity score including calculating the cosine similarity, the cosine similarity is calculated by the following formula: ; When calculating the similarity score and the Euclidean distance, the Euclidean distance is calculated by the following formula: .

[0056] The present application also provides a rental contract intelligent matching recommendation device based on personal portraits, the intelligent matching recommendation device comprising: A collection module 401, the collection module 401 is used to collect subjective information and objective information of the target user; the subjective information is information collected through the active behavior of the target user, and the objective information is information collected through the passive behavior of the target user; A user portrait building module 402, the user portrait building module 402 is used to build a user portrait of the target user based on the subjective information and objective information of the target user; A feature description acquisition module 403, the feature description acquisition module 403 is used to acquire a feature description of the leased object based on the attributes of the leased object; A similarity score calculation module 404, which is used to calculate a similarity score between the user portrait of the target user and the feature description of the leased item; The recommendation module 405 is used to select the rental item with the highest similarity as the recommendation result according to the similarity score.

[0057] Those skilled in the art will appreciate that all or part of the functions of the various methods in the above-mentioned embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, and the storage medium can include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to implement the above-mentioned functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above-mentioned functions can be implemented. In addition, when all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can also be stored in a storage medium such as a server, another computer, disk, optical disk, flash disk or mobile hard disk, and can be downloaded or copied and saved in the memory of the local device, or the system of the local device is updated, and when the program in the memory is executed by the processor, all or part of the functions in the above-mentioned embodiments can be implemented.

[0058] The above specific examples are used to illustrate the present application, which is only used to help understand the present application and is not intended to limit the present application. For technicians in the technical field to which the present application belongs, they can also make some simple deductions, deformations or substitutions based on the ideas of the present application.

Claims

1. A rental contract intelligent matching recommendation method based on personal portrait, characterized in that: The recommended methods include: Collecting and obtaining subjective information and objective information of a target user; the subjective information is information collected through the active behavior of the target user, and the objective information is information collected through the passive behavior of the target user; Based on the subjective information and objective information of the target user, establish a user profile of the target user; Based on the attributes of the leased object, obtaining a characteristic description of the leased object; Calculate the similarity score between the user profile of the target user and the feature description of the leased item; Based on the similarity score, the rental item with the highest similarity is selected as the recommendation result.

2. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 1, characterized in that: The collecting and obtaining of the subjective information and objective information of the target user includes: Collecting the subjective information of the target user through the active behavior of the target user; the initiative includes but is not limited to: questionnaire survey and / or actively filling in information; Objective information of the target user is collected through the passive behavior of the target user; the passive behavior includes but is not limited to: behavior data, and / or transaction records, and / or location service data, and / or credit records.

3. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 1, characterized in that: The step of establishing a user profile of the target user based on the subjective information and objective information of the target user includes: Based on the subjective information of the target user, a subjective information vector group of the target user is obtained; Based on the objective information of the target user, obtaining an objective information vector group of the target user; Based on the subjective information vector group and the objective information vector group, a user profile of the target user is constructed. .

4. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 3, characterized in that: The subjective information vector group includes a subjective information vector of at least one dimension, and the objective information vector group includes an objective information vector of at least one dimension; The dimensions of the subjective information vector group and the objective information vector group are consistent.

5. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 4, characterized in that: User profile of the target user It is expressed by the following formula: ； in, The target user User portraits, The target user The subjective information vector group, The target user Objective information vector group.

6. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 5, characterized in that: The acquiring of a characteristic description of the leased object based on the attribute of the leased object includes: The characteristic description of the leased property is expressed by the following formula : ； Subjective information vector group ; Objective information vector group ; in, represents the k-dimensional feature vector, represents the k-dimensional subjective information vector, Represents the k-dimensional objective information vector.

7. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 6, characterized in that: The calculating and obtaining the similarity score between the user portrait of the target user and the feature description of the leased item includes: Based on the properties of the leased object, different subjective information weights and objective information weights are configured for different leased objects; Obtaining a specific portrait of the target user; the specific portrait is a weighted combination of the subjective information weight and the objective information weight of the user portrait; Calculating a similarity score between a specific profile of the target user and a feature description of the rental item; Wherein, calculating the similarity score includes calculating cosine similarity or Euclidean distance.

8. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 7, characterized in that: The specific image is represented by the following formula: ； in, Specific portraits of target users, is the subjective information weight, is the objective information weight.

9. The method for intelligent matching and recommendation of lease contracts based on personal portraits as claimed in claim 8, characterized in that: When calculating the similarity score including calculating the cosine similarity, the cosine similarity is calculated by the following formula: ； When calculating the similarity score and the Euclidean distance, the Euclidean distance is calculated by the following formula: 。 10. A rental contract intelligent matching recommendation device based on personal portrait, characterized in that: The intelligent matching recommendation device comprises: A collection module, the collection module is used to collect subjective information and objective information of the target user; the subjective information is information collected through the active behavior of the target user, and the objective information is information collected through the passive behavior of the target user; A user portrait building module, the user portrait building module is used to build a user portrait of the target user based on the subjective information and objective information of the target user; A feature description acquisition module, the feature description acquisition module is used to acquire a feature description of the leased object based on the attributes of the leased object; A similarity score calculation module, the similarity score calculation module is used to calculate a similarity score between a user portrait of a target user and a feature description of the leased item; The recommendation module is used to select the rental item with the highest similarity as the recommendation result according to the similarity score.

Images