User demand generation method based on user portrait

Abstract

The invention discloses a user demand generation method based on a user portrait, and the method comprises the following steps: S1, obtaining user data from an online platform, and building the user portrait according to the user data; s2, based on the established user portrait, obtaining a demand directly presented by the user, which is called as a presentation demand; s3, on the basis of the user portrait and the presentation demand of the user, further obtaining an extension demand and an optimization demand of the user through an AI model; and S4, based on the acquired various demands of the user, realizing accurate pushing aiming at the user demands on an online platform in combination with the user portrait. According to the method, based on the established user portrait, the user is taken as a core, and the full link of demand mining is deeply focused, so that accurate insight and dynamic updating of the demand are realized; on this basis, the mined real demand is converted into a basis for accurate pushing, so that the user can feel the understood value in efficient acquisition; meanwhile, potential interest points of the users can be mined, and potential users can be converted.

Description

Technical Field

[0001] This invention relates to the field of business model technology, and in particular to a method for generating user needs based on user profiles. Background Technology

[0002] With the advent of the intelligent era, emerging technologies such as artificial intelligence (AI), machine learning, the Internet of Things (IoT), and blockchain are beginning to transform business models, especially in areas such as smart manufacturing, unmanned delivery, and intelligent customer service. This is making the consumer experience more intelligent, simplifying shopping and daily life through smart devices, voice assistants, and other tools.

[0003] However, existing app user profiling and recommendation systems still rely on tagging users, describing them using numerous tags. This tagged user profile fails to resonate with individual user personalities, only capturing explicit user needs. Furthermore, it relies on aggressive recommendations to convert potential users, resulting in low conversion rates, user resistance, and recommendations that are often disconnected from actual user needs, negatively impacting user experience and satisfaction. Summary of the Invention

[0004] The purpose of this invention is to provide a user needs generation method based on user profiles, in order to solve the technical problems existing in the background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for generating user needs based on user profiles includes the following steps:

[0007] S1. Obtain user data from online platforms and build user profiles accordingly;

[0008] S2. Based on the established user profile, obtain the user's directly presented needs, which are called presentation needs;

[0009] S3. Based on user profiles and user presentation needs, further obtain users' extended and optimization needs through AI models;

[0010] S4. Based on the various user needs obtained and combined with user profiles, accurate push notifications are made to users on the online platform.

[0011] Furthermore, in step S1, the process of creating a user profile specifically includes the following steps:

[0012] S11. Collect basic user data on the online platform and collect user behavior data in real time;

[0013] S12. Based on the user's basic data and behavioral data, establish and continuously update the user's mental model. The mental model is used to reflect the user's cognitive and behavioral tendencies.

[0014] S13. Combine user basic data, behavioral data, and mental models to form a user profile.

[0015] Furthermore, in step S2, when obtaining the user's presentation requirements, on the one hand, the user's needs are directly carried over as the user's presentation requirements based on the needs actively expressed by the user on the online platform;

[0016] On the other hand, based on the basic data in the user profile, AI models can be used to uncover the needs that users may express, which can then be used as the user's presentation needs.

[0017] Furthermore, in step S3, when obtaining the user's extended needs, the scenarios include, but are not limited to, the following:

[0018] Based on the basic data in user profiles, combined with behavioral data and / or mental models in user profiles, new demands are generated through AI models.

[0019] Based on a user's individual presentation requirements, new requirements different from the original requirements are generated through AI models;

[0020] Based on a user's individual presentation needs, combined with user profiles, new needs that differ from the original needs are generated through AI models;

[0021] Based on multiple user presentation needs, new needs are generated through AI models;

[0022] Based on multiple user presentation needs and combined with user profiles, new needs are generated through AI models;

[0023] In step S4, extended requirements are pushed to provide users with more options.

[0024] Furthermore, in step S3, when obtaining the user's optimization requirements, the AI ​​model directly optimizes based on the user's presentation requirements, or optimizes based on the user profile, and generates new optimization requirements through the AI ​​model.

[0025] In step S4, optimization requests are pushed to recommend better options to users.

[0026] Furthermore, in step S11, psychological characteristics are predefined, and the association configuration between psychological characteristics and users' basic data and behavioral data is established. Corresponding collection methods are configured for each type of psychological characteristic, and psychological characteristic type evaluation criteria and psychological characteristic scoring criteria are configured.

[0027] In step S12, when establishing and continuously updating the user's mental model:

[0028] The collected basic user data and behavioral data are collectively referred to as user data. First, the user data is cleaned to remove noisy data, and then the data is standardized and completed.

[0029] Then, feature extraction is performed on the processed user data, keywords are extracted, and combined with sentiment analysis algorithms, the user's emotional tendencies and values ​​are identified;

[0030] Next, based on the preset association configuration in step S11, the psychological features corresponding to the user data after feature extraction are determined, and the psychological features are quantified to quantify their corresponding psychological feature scores.

[0031] User data is categorized according to different psychological characteristics, and the categorized user data is labeled to clarify the psychological characteristics reflected by each piece of user data and its score.

[0032] Based on the scores of all the user's psychological characteristics, a comprehensive feature vector is obtained, and a psychological model of the user is generated accordingly.

[0033] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0034] This invention provides a user demand generation method based on user profiles. Based on the established user profiles, it focuses on the user as the core and deeply focuses on the entire process of demand mining to achieve accurate insight and dynamic updates of demands. On this basis, the mined real demands are transformed into the basis for precise push notifications, allowing users to feel understood and valued while efficiently acquiring information. At the same time, it can also discover users' potential interests and convert potential users. Attached Figure Description

[0035] Figure 1 This is a flowchart illustrating a user requirement generation method based on user profiles provided by the present invention. Detailed Implementation

[0036] To make the technical means, creative features, objectives and effects of this invention easier to understand, the following description, in conjunction with the accompanying drawings and specific embodiments, further explains how this invention is implemented.

[0037] In one specific embodiment, refer to Figure 1 As shown, a user requirement generation method based on user profiles includes the following steps:

[0038] S1. Obtain user data from online platforms and build user profiles accordingly.

[0039] S2. Based on the established user profile, obtain the user's directly presented needs, which are called presentation needs.

[0040] S3. Based on user profiles and user presentation needs, further obtain users' extended and optimization needs through AI models.

[0041] S4. Based on the various user needs obtained and combined with user profiles, accurate push notifications are made to users on the online platform.

[0042] In one specific embodiment:

[0043] In step S1, user data is obtained from the online platform, and user profiles are built based on this data. This includes the following steps:

[0044] S11. Collect basic user data on the online platform and collect user behavior data in real time.

[0045] Traditional user profiles only include basic data such as gender, age, address, and occupation, failing to capture dynamic behavior and deep psychological motivations. This embodiment also collects user behavior data, including user clicks, dwell time, social interactions, and consumption patterns. Data collection can be achieved through web crawling systems and / or point-based data collection. Data from different sources and of different types is integrated to form a complete user behavior dataset.

[0046] In addition, psychological characteristics are predefined for subsequent psychological model building. These models reflect users' cognition (such as cognitive preferences, decision-making rationality, and value orientation) and behavioral tendencies (such as impulsivity, social desire, exploratory desire, and stability). Furthermore, a configuration is established to link psychological characteristics with users' basic data and behavioral data. Corresponding data collection methods are configured for each type of psychological characteristic, along with psychological characteristic type evaluation criteria and scoring standards.

[0047] When defining psychological characteristics, they can be categorized into multiple levels. Higher levels may include human nature, preferences, principles, intelligence, and morality; through these characteristics, one can gain a comprehensive understanding of a person's cognitive and behavioral tendencies. Lower levels, such as human nature, may include basic psychological traits such as kindness, ego, curiosity, competitiveness, willingness to cooperate, social skills, calmness, tension, optimism, and pessimism.

[0048] In this embodiment, some psychological characteristics are obtained in the following ways:

[0049] Kindness is manifested in caring for others, being helpful, and being considerate. Keywords related to caring, being considerate, and supporting others can be extracted from users' words and actions on the platform.

[0050] Self: This manifests as egocentrism, self-expression, and independence. It can be assessed by analyzing the frequency with which users mention their own interests and opinions during interactions.

[0051] Curiosity: This manifests as interest in and desire to explore new things and new knowledge. It can be assessed through a user's search history and the frequency with which they browse different types of content.

[0052] Competitive psychology: This manifests as a competitive spirit and desire to win in various activities. It is assessed through users' performance in activities such as games and discussions.

[0053] Willingness to cooperate: This refers to the user's willingness and ability to cooperate with others. It is assessed through the user's participation and cooperative behavior in group activities.

[0054] The definition and subdivision of each of these psychological characteristics lay the foundation for subsequent data collection, analysis, and processing. These characteristics help understand user behavior and preferences and provide essential information for building accurate user profiles. Through comprehensive analysis and evaluation of these psychological characteristics, we can more accurately predict user behavior in different situations, identify user needs, and provide personalized content recommendation services.

[0055] S12. Based on users' basic data and behavioral data, establish and continuously update users' mental models. Specifically:

[0056] The collected basic and behavioral data of users are collectively referred to as user data. The user data is first cleaned to remove noise (such as duplicate content and meaningless comments). The data is then standardized to conform to a unified standard and format for easier subsequent analysis and processing. For example, data from different platforms is converted to the same time format and text encoding. In some cases, data gaps may exist; in such cases, data completion is performed. Missing data can be supplemented through inference or by sourcing from other data sources to improve data completeness.

[0057] Then, feature extraction is performed on the processed user data to extract keywords, and combined with sentiment analysis algorithms to identify users' emotional tendencies and values. Natural Language Processing (NLP) techniques can be used to extract key feature words, sentiment tendencies, and semantic information from users' statements. For example, sentiment analysis can determine a user's emotional state, and keyword extraction can identify a user's interests and concerns.

[0058] Next, based on the preset association configuration in step S11, the psychological features corresponding to the extracted user data are determined, and these psychological features are quantified to obtain their corresponding scores. Evaluation criteria can be set for each psychological feature, and machine learning and natural language processing techniques can be used to transform user data into psychological feature scores. For example, scores can be awarded by calculating the frequency of keyword occurrences, the frequency and intensity of user behaviors, etc. If the frequency of certain keywords exceeds a preset value, the score for that psychological feature is higher. Sentiment analysis algorithms can also be used to score the positivity and negativity of user comments.

[0059] User data is categorized according to different psychological characteristics, and the categorized user data is labeled to clarify the psychological characteristics reflected by each piece of user data and its score. For example, words and behaviors involving kindness are categorized into one group, and behaviors involving competitive psychology are categorized into another group. The categorized data is labeled to clarify the psychological characteristics reflected by each piece of data and its score. For example, a user comment is labeled as "caring about others" and assigned a certain weight score to reflect its contribution to the characteristic of kindness.

[0060] Based on the user's scores for all psychological characteristics, a comprehensive feature vector is obtained, from which a psychological model of the user is generated. Furthermore, multi-dimensional analysis of the user's psychological characteristics can be performed, considering the interrelationships and influences between different characteristics. For example, a user's "social skills" may affect their "willingness to cooperate," and "calmness" may be negatively correlated with "nervousness."

[0061] S13. Combine user basic data, behavioral data, and mental models to form a user profile.

[0062] Building upon individual user profile models, group profiles can be constructed to identify user groups with similar psychological characteristics and behavioral patterns. For example, cluster analysis can identify a group of users with a high degree of kindness and willingness to cooperate.

[0063] In step S2, based on the established user profile, the user's direct presentation needs are obtained, which are referred to as presentation needs.

[0064] When acquiring user presentation requirements, on the one hand, based on the needs actively expressed by users on the online platform, these needs are directly incorporated as the user's presentation requirements. For example, if a user searches for a product, then purchasing that product is directly presented as the user's presentation requirement.

[0065] On the other hand, based on the basic data in user profiles, AI models can be used to uncover the needs that users may express, and these needs can be presented as user-defined requirements. For example, based on a user's profession, essential products for that profession can be identified, and purchasing those products can be presented as a user-defined requirement.

[0066] In step S3, based on the user profile and the user's presentation needs, the AI ​​model is used to further obtain the user's extended needs and optimization needs.

[0067] When acquiring users' extended needs, this includes, but is not limited to, the following scenarios:

[0068] Based on the fundamental data in user profiles, combined with behavioral data and / or mental models from those profiles, new demands can be generated through AI models. For example, if a user profile indicates that they frequently stay up late and are a programmer, an AI model could determine that the user might need liver-protecting health supplements, thus generating an extended demand to purchase these supplements.

[0069] Based on a user's single presented need, an AI model can be used to generate new needs that differ from the original needs. For example, if a user's presented need is to buy an airline ticket, the AI ​​model can then generate an extended need to buy travel insurance.

[0070] Based on a user's individual stated needs, and combined with user profiles, AI models can be used to generate new needs that differ from the original needs. For example, if a user's stated need is to buy sportswear, and the user profile reveals that the user loves sports, the AI ​​model can then generate an extended need for a gym membership.

[0071] Based on multiple presented user needs, AI models can stimulate the generation of new needs. These generated extended needs can supplement existing needs; for example, if a user's presented needs are to buy baby formula and toys, the AI ​​model can generate an extended need to obtain childcare services. Alternatively, these extended needs can substitute for existing needs; for instance, if a user's presented need is to buy two different products, and a different high-end product exists that combines the functions of both, the AI ​​model can generate an extended need to purchase that high-end product.

[0072] Based on multiple presented user needs, and combined with user profiles, AI models can stimulate the generation of new needs. For example, if a user's presented needs are to purchase a gym membership and buy health supplements, and the user profile reveals that the user is middle-aged or elderly, the AI ​​model can then generate an extended need for health check-up services.

[0073] Furthermore, when acquiring user optimization needs, the AI ​​model directly optimizes based on the user's presentation requirements, or optimizes based on the user profile, and generates new optimization needs through the AI ​​model. Optimization can be in multiple aspects, such as making the price cheaper, or making the price more expensive but with better performance; both fall under the category of optimization. The specific direction of optimization can be determined by combining user profiles.

[0074] For example, if a user's stated need is to buy a mobile phone, and there is a lack of corresponding user profile data, the AI ​​model can generate two optimization needs simultaneously: buying a cheaper mobile phone and buying a high-end mobile phone. However, if the user profile data indicates that the user prefers to buy high-end products, the AI ​​model will only generate the optimization need to buy a high-end mobile phone.

[0075] For example, if a user's stated need is to purchase a course, and user profile data indicates that the user is introverted, then the AI ​​model will generate an optimized need to purchase online courses rather than offline courses.

[0076] For example, if a user's stated need is to purchase a certain product, and user profile data indicates that the user has a high psychological impulse score, then the AI ​​model can generate an optimized need to purchase a limited-time discounted product.

[0077] In step S4, based on the acquired user needs and user profiles, precise push notifications are implemented on the online platform to meet those needs. For example, when making precise push notifications based on user needs, location information from the user profile needs to be considered. If the user's need is food delivery service, then food delivery services for the corresponding city and address will be pushed. Extending the push notifications to provide users with more options, and optimizing the push notifications to recommend better options to users, further expands the push notifications.

[0078] In summary, the user demand generation method based on user profiles provided by this invention, with users at its core, deeply focuses on the entire chain of demand mining, based on the established user profiles, to achieve accurate insight and dynamic updates of demands; on this basis, the mined real demands are transformed into the basis for accurate push notifications, allowing users to feel the value of being understood while efficiently acquiring information; at the same time, it can also discover users' potential interests and convert potential users.

[0079] Finally, it should be noted that the above description is only an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for generating user needs based on user profiles, characterized in that, Includes the following steps: S1. Obtain user data from online platforms and build user profiles accordingly; S2. Based on the established user profile, obtain the user's directly presented needs, which are called presentation needs; S3. Based on user profiles and user presentation needs, further obtain users' extended and optimization needs through AI models; S4. Based on the various user needs obtained and combined with user profiles, accurate push notifications are made to users on the online platform.

2. The user requirement generation method based on user profiles according to claim 1, characterized in that, In step S1, the process of creating a user profile includes the following steps: S11. Collect basic user data on the online platform and collect user behavior data in real time; S12. Based on the user's basic data and behavioral data, establish and continuously update the user's mental model. The mental model is used to reflect the user's cognitive and behavioral tendencies. S13. Combine user basic data, behavioral data, and mental models to form a user profile.

3. The user requirement generation method based on user profiles according to claim 2, characterized in that, In step S2, when obtaining the user's presentation requirements, on the one hand, the user's needs are directly carried over to the online platform as the user's presentation requirements; On the other hand, based on the basic data in the user profile, AI models can be used to uncover the needs that users may express, which can then be used as the user's presentation needs.

4. The user requirement generation method based on user profiles according to claim 3, characterized in that, In step S3, when obtaining the user's extended needs, the following scenarios are included but not limited to: Based on the basic data in user profiles, combined with behavioral data and / or mental models in user profiles, new demands are generated through AI models. Based on a user's individual presentation requirements, new requirements different from the original requirements are generated through AI models; Based on a user's individual presentation needs, combined with user profiles, new needs that differ from the original needs are generated through AI models; Based on multiple user presentation needs, new needs are generated through AI models; Based on multiple user presentation needs and combined with user profiles, new needs are generated through AI models; In step S4, extended requirements are pushed to provide users with more options.

5. The user requirement generation method based on user profiles according to claim 4, characterized in that, In step S3, when obtaining the user's optimization requirements, the AI ​​model directly optimizes based on the user's presentation requirements, or optimizes based on the user profile, and generates new optimization requirements through the AI ​​model. In step S4, optimization requests are pushed to recommend better options to users.

6. The user requirement generation method based on user profiles according to claim 5, characterized in that, In step S11, psychological characteristics are predefined, and the association configuration between psychological characteristics and users' basic data and behavioral data is established. Corresponding collection methods are configured for each type of psychological characteristic, and psychological characteristic type evaluation criteria and psychological characteristic scoring criteria are configured. In step S12, when establishing and continuously updating the user's mental model: The collected basic user data and behavioral data are collectively referred to as user data. First, the user data is cleaned to remove noisy data, and then the data is standardized and completed. Then, feature extraction is performed on the processed user data, keywords are extracted, and combined with sentiment analysis algorithms, the user's emotional tendencies and values ​​are identified; Next, based on the preset association configuration in step S11, the psychological features corresponding to the user data after feature extraction are determined, and the psychological features are quantified to quantify their corresponding psychological feature scores. User data is categorized according to different psychological characteristics, and the categorized user data is labeled to clarify the psychological characteristics reflected by each piece of user data and its score. Based on the scores of all the user's psychological characteristics, a comprehensive feature vector is obtained, and a psychological model of the user is generated accordingly.

Images