User service demand intelligent evaluation system based on portrait construction

Abstract

The invention discloses a user service demand intelligent evaluation system based on portrait construction, relates to the technical field of user service demand evaluation, and solves the technical problem that in the prior art, a portrait label system cannot be constructed according to portrait label setting, so that an evaluation standard cannot be formulated in real time for targeted service demand evaluation. Specifically, a data dimension detection unit carries out dimension analysis on data acquisition in a user service demand evaluation process; the demand prediction lag analysis unit is used for carrying out demand prediction lag analysis on user dimension data acquisition; the data screening module is used for screening the collected dimension data and dividing the dimension data; the portrait updating module is used for carrying out portrait label setting on the acquired dimension data of the user, carrying out weight setting on portrait labels, and constructing a portrait label system according to the portrait labels with different weights; and the service demand evaluation unit is used for intelligently evaluating the service demand of the user according to the portrait label system.

Description

Technical Field

[0001] The present invention relates to the technical field of user service demand assessment, and specifically to an intelligent assessment system for user service demands based on portrait construction. Background Art

[0002] The intelligent assessment of user service demands based on portrait construction refers to the process of integrating multi-dimensional information such as user basic data, behavior data, and historical service records to construct a dynamic user portrait, and using technologies such as machine learning, natural language processing, and data analysis to achieve accurate identification, priority ranking, and personalized response to users' potential service demands.

[0003] However, in the prior art, the data dimension is single during the assessment of user service demands, making it difficult to capture users' dynamic behavior characteristics; and the demand prediction lags behind, unable to respond to scenario changes in real time; at the same time, it is impossible to collect and screen dimension data, and the impact of misoperations cannot be excluded; in addition, it is impossible to construct a portrait tag system according to the portrait tag settings, so that it is impossible to formulate an assessment standard in real time for targeted service demand assessment.

[0004] In view of the above technical defects, a solution is proposed now. Summary of the Invention

[0005] The purpose of the present invention is to solve the above-mentioned problems, and to propose an intelligent assessment system for user service demands based on portrait construction.

[0006] The purpose of the present invention can be achieved by the following technical solutions: An intelligent assessment system for user service demands based on portrait construction includes a service demand assessment platform, and the service demand platform includes: A data dimension detection unit for performing dimension analysis on data collection during the assessment of user service demands; A demand prediction lag analysis unit for performing demand prediction lag analysis on the collection of user dimension data; After the data dimension detection unit determines that the dimension detection meets the actual requirements, a data screening module screens the collected dimension data and divides each dimension data; A portrait update module sets portrait tags for the collected user dimension data, sets weights for the portrait tags, and constructs a portrait tag system according to the portrait tags with different weights; A service demand assessment unit for performing intelligent assessment of user service demands according to the portrait tag system.

[0007] As a preferred implementation manner of the present invention, the operation process of the data dimension detection unit is as follows: Collect user service demand data, where the service demand is a commodity purchase demand or a consulting service demand; perform dimension detection on the currently collected data, where the data dimensions are divided into user dimensions, behavior dimensions, and sentiment dimensions; when facing different service demands, if there are different numerical manifestations in any dimension data of the user, it is determined that the currently collected data dimension meets the actual evaluation requirements; If there are no different numerical manifestations in any dimension data of the user, perform a single dimension evaluation; When the user generates dimension data, determine the static and dynamic collection of the user data. Statistically analyze the ratio of the number of static collections to the number of dynamic collections during the collection of each dimension data currently. If the ratio exceeds the set ratio threshold, or the ratio continues to increase, it is inferred that as the user's dimension data increases, the current data dimension is single, and the satisfaction of the dimension data detection requirements shows a downward trend; if the ratio does not exceed the set ratio threshold and does not continue to increase, it is inferred that as the user's dimension data increases, the current data dimension meets the user demand evaluation.

[0008] As a preferred embodiment of the present invention, the operation process of the demand prediction lag analysis unit is as follows: Divide the user dimension data collection stage into several sub-stages, collect the floating trend of the dimension data in each sub-stage and the floating trend of the dimension data in the entire collection stage. If the corresponding floating trends are inconsistent, mark the sub-stage as a floating stage; if the corresponding floating trends are consistent, mark the sub-stage as a stable stage; During the floating stage, collect the interval duration between the moment of analyzing the trend of the user dimension data and the end moment of the floating stage; during the stable stage, collect the numerical increase span of the corresponding dimension data in multiple stable stages when analyzing the trend of the user dimension data; When the interval duration within the floating stage exceeds the interval duration threshold, or the numerical increase span within the stable stage exceeds the span threshold, it is inferred that the lag analysis during the dimension data collection process is abnormal, and the service demand evaluation platform shortens the prediction period of the dimension data collection, that is, perform floating real-time prediction during the floating stage, and set a critical value for the numerical floating span within the stable stage, and perform demand prediction in a timely manner if it exceeds; when the interval duration within the floating stage does not exceed the interval duration threshold and the numerical increase span within the stable stage does not exceed the span threshold, it is inferred that the lag analysis during the dimension data collection process is normal.

[0009] As a preferred embodiment of the present invention, the operation process of the data screening module is as follows: Perform generation method classification according to the analysis of the generation time of the dimension data, that is, divide it into active generation and passive generation; Collect the number of times of active generation and passive generation of users. If the dimensional data of the type corresponding to the number of times of active generation continuously generates data without interruption during the data collection phase, mark the corresponding dimensional data as continuously concerned data; if the dimensional data of the type corresponding to the number of times of active generation is generated in a concentrated period during the data collection phase and is not continuously generated, mark the corresponding dimensional data as briefly concerned data; If the ratio of the interval duration between consecutive generation times of the dimensional data of the type corresponding to the number of times of passive generation to the duration of continuous generation of the dimensional data is lower than the duration ratio threshold, mark the corresponding dimensional data as data concerned due to push; if the ratio of the interval duration between consecutive generation times of the dimensional data of the type corresponding to the number of times of passive generation to the duration of continuous generation of the dimensional data is not lower than the duration ratio threshold, mark the corresponding dimensional data as data pushed but not concerned.

[0010] As a preferred embodiment of the present invention, the operation process of the portrait update module is as follows: Set portrait labels for each type of dimensional data, and the portrait labels are divided into risk preference labels, category preference labels, and demand preference labels; perform feature vector transformation on each type of label, and the feature vector transformation is to select one within the range of each type of label for feature vector transformation; Among them, the feature vector of the risk preference label is the excess amount of the service demand usage period lower than the floating frequency threshold compared to the user's demand usage period; the category preference label represents the overlapping browsing frequency of the service demand types in the dimensional data; the demand preference label represents the continuous browsing frequency of the user within the service demand usage period obtained.

[0011] As a preferred embodiment of the present invention, after completing the feature vector transformation, construct a portrait label system for the corresponding data of each type of portrait label, and obtain the ratio of the feature vector value of the corresponding same-type label to the set threshold of the corresponding feature vector according to the service demand historically selected by the user, and mark it as the portrait evaluation ratio; compare the portrait evaluation ratio of each type of label in real time with the portrait evaluation ratio of the historical same-type label. Whether the portrait evaluation ratio of each type of label in real time exceeds the portrait evaluation ratio corresponding to the historically selected service demand is used as the weight setting standard for the current portrait label, that is, the three portrait labels are respectively set as high weight, medium weight, and low weight, and a portrait label system is constructed according to the weights.

[0012] As a preferred embodiment of the present invention, the operation process of the service demand evaluation unit is as follows: According to the portrait label system, data analysis is carried out on each weighted portrait label in the system. The peak increase span of the eigenvector values corresponding to the high-weight portrait labels and the valley decrease span of the eigenvector values corresponding to the low-weight portrait labels in the portrait label system are collected, and the span value ratio is calculated according to the ratio. At the same time, in the floating stage of the eigenvectors corresponding to the medium-weight portrait labels in the portrait label system, the frequency value ratio of the range from the increase to the high-weight eigenvector value and the range from the decrease to the low-weight eigenvector value is obtained.

[0013] As a preferred embodiment of the present invention, if the span value ratio exceeds the span value ratio threshold, it is inferred that the two-pole weight division of the current portrait label system fits, and the two-pole weights of the current portrait label system are used as the user service demand evaluation criteria; if the span value ratio does not exceed the span value ratio threshold, it is inferred that the two-pole weight division of the current portrait label system does not fit, and the floating trend of the two-pole weights corresponding to the current portrait label system is used as the user service demand evaluation criteria; If the frequency value ratio exceeds the maximum value of the frequency value ratio range, it is inferred that there is an increase in the data of the high-weight portrait labels in the current portrait label system, and data analysis is carried out based on the increased portrait labels to evaluate the service demand; if the frequency value ratio does not exceed the minimum value of the frequency value ratio range, it is inferred that there is an increase in the data of the low-weight portrait labels in the current portrait label system, and data analysis is carried out based on the increased portrait labels to evaluate the service demand; If the frequency value ratio is within the frequency value ratio range, it is inferred that the floating impact of the weight data in the current portrait label system is small, and the current portrait label system is used as the user service demand evaluation criteria.

[0014] As a preferred embodiment of the present invention, after determining the user service demand evaluation criteria, user evaluation is carried out. According to the weights of each portrait label in the portrait label system, combined with the influence of the corresponding dimension data, the service demand of the current user is inferred.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the present invention, detection and collection are carried out during the detection of user service data, ensuring high availability of the collected data of users during portrait construction, and avoiding the deviation of user service demand evaluation caused by the lag or incomplete data of the collected data; a lag analysis of the demand prediction for the collection of user dimension data is carried out. By analyzing the process of dimension data collection, it is inferred whether there is a lag in the collected data when the current dimension data is used as a user demand evaluation parameter, resulting in untimely real-time demand evaluation, a decrease in the accuracy of user service demand evaluation, and an inability to accurately predict in line with the actual demand.

[0016] 2. In the present invention, the collected dimensional data is screened to avoid incorrect operations by users when the dimensional data is generated. Due to personal habits or the influence of the advertisement push position on the amount of dimensional data collected, errors may occur in the evaluation of user needs, affecting the accuracy of the evaluation results. Portrait labels are set for the collected dimensional data of users, and feature vector conversion is performed according to the portrait labels. Through the setting of portrait labels and the corresponding feature vector analysis, the portrait labels are updated in real time to facilitate the evaluation of user needs based on the latest real-time portrait labels, thereby improving the accuracy of user need evaluation and avoiding excessive dimensional data corresponding to multiple portrait labels, resulting in a large amount of data processing work for service need evaluation and prone to errors. By updating the portrait to obtain the latest portrait labels, the availability of the existing collected data and the accuracy of analysis are maximally ensured.

[0017] 3. In the present invention, the service needs of users are intelligently evaluated according to the portrait label system, and the service needs are evaluated according to the real-time updated portrait label system, effectively improving the accuracy of user evaluation and ensuring the availability of real-time collected data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings.

[0019] Figure 1 is the principle block diagram of Embodiment 1 of the present invention; Figure 2 is the principle block diagram of Embodiment 2 of the present invention; Figure 3 is the principle block diagram of Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0021] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0022] Embodiment 1 Please refer toFigure 1 As shown, an intelligent evaluation system for user service requirements based on portrait construction includes a service requirement evaluation platform, where the service requirement evaluation platform includes a data dimension detection unit and a demand prediction lag analysis unit; this embodiment is used for detection and collection during user service data detection to ensure high availability of the collected data of users during portrait construction, and avoid deviation in user service requirement evaluation caused by lag or incomplete data in the collected data. The service requirement evaluation platform generates a data dimension detection signal and sends it to the data dimension detection unit. After receiving the data dimension detection signal, the data dimension detection unit performs dimension analysis on the data collection during the user service requirement evaluation to ensure comprehensive real-time data collection dimensions, overcome the influence of single collected data, and improve the accuracy of user service requirement evaluation. Perform user service requirement data collection, where the service requirement is a commodity purchase requirement or a consulting service requirement; perform dimension detection on the currently collected data, where the data dimensions are divided into user dimension, behavior dimension, and emotion dimension; among them, the user dimension is represented by the user's age, location, consumption level, etc.; the behavior dimension is represented by the user's web browsing record, interaction duration, click heat map, etc.; the emotion dimension is represented by customer service dialogue text, evaluation semantic analysis. When facing different service requirements, if there are different numerical manifestations in any dimension data of the user, it is determined that the currently collected data dimension meets the actual evaluation requirements. For example, the age in the user dimension data does not overlap with the target group of the current service requirement, the number of browsing records corresponding to the service requirement in the behavior dimension does not increase, and the continuous frequency of the dialogue text in the emotion dimension gradually decreases. If there are no different numerical manifestations in any dimension data of the user, perform a single dimension evaluation. When the user generates dimension data, determine the static and dynamic collection of user data. Static collection means that the user records in a timely manner after generating the current dimension data, such as recording after the user's web browsing record increases; dynamic collection means that the user continues to analyze and record with the current dimension data after generating the current dimension data. For example, after the user's web browsing record increases and is recorded, dynamically record the browsing track related to the user's current web browsing record, and the related browsing track corresponds to the same type of commodity or the same type of service. Statistically analyze the ratio of the number of times of static collection and dynamic collection during the current collection of each dimension data. If the ratio of the number of times exceeds the set ratio threshold, or the ratio of the number of times continues to increase, it is inferred that as the user's dimension data increases, the current data dimension is single, and the satisfaction of the dimension data detection requirement shows a downward trend; resulting in the inability to perform user service requirement evaluation on the current dimension data. If the ratio of times does not exceed the set ratio threshold and does not continue to increase, it is inferred that as the user's dimensional data increases, the current data dimension meets the user demand assessment; The dimensional data in this part is limited to the behavior dimension and the emotion dimension. The behavior dimension and the emotion dimension can reflect the initiative of the user's service demand; the service demand assessment platform adjusts the dimensional data collection type according to the dimensional detection result; After completing the dimensional data detection, the service demand assessment platform generates a demand prediction lag analysis signal and sends it to the demand prediction lag analysis unit; After receiving the demand prediction lag analysis signal, the demand prediction lag analysis unit conducts a demand prediction lag analysis on the user's dimensional data collection. By inferring through the dimensional data collection analysis process, when the current dimensional data is used as a user demand assessment parameter, whether there is a lag in data collection that causes untimely real-time demand assessment, resulting in a decline in the accuracy of user service demand assessment and an inability to accurately predict according to actual needs; The user dimensional data collection stage is divided into several sub-stages, and the floating trends of the dimensional data in each sub-stage and the floating trend of the dimensional data in the entire collection stage are collected. If the corresponding floating trends are inconsistent, the sub-stage is marked as a floating stage; if the corresponding floating trends are consistent, the sub-stage is marked as a stable stage; During the floating stage, the time interval between the moment of collecting the user dimensional data trend analysis and the end moment of the floating stage is collected; during the stable stage, the numerical increase span of the corresponding dimensional data in multiple stable stages is collected during the user dimensional data trend analysis; When the time interval duration within the floating stage exceeds the interval duration threshold, or the numerical increase span within the stable stage exceeds the span threshold, it is inferred that there is an abnormality in the lag analysis during the dimensional data collection process. The service demand assessment platform shortens the dimensional data collection prediction cycle, that is, performs floating real-time prediction within the floating stage, and sets a critical value for the numerical floating span within the stable stage, and conducts demand prediction in a timely manner if it exceeds; When the time interval duration within the floating stage does not exceed the interval duration threshold, and the numerical increase span within the stable stage does not exceed the span threshold, it is inferred that the lag analysis during the dimensional data collection process is normal; The thresholds used in this embodiment are all parameters artificially set by those skilled in the art during actual operation, and are used for the detection of various types of collected data; Embodiment 2 Based on the previous embodiment, this embodiment improves the service demand assessment platform and makes further processing after dimensional detection and lag analysis. Please refer to Figure 2 As shown, the data dimension detection unit is communicatively connected to a data screening module; the demand prediction lag analysis unit is communicatively connected to a portrait update module; After determining that the dimension detection meets the actual requirements, the data dimension detection unit generates a data screening signal and sends it to the data screening module; After receiving the signal, the data screening module screens the collected dimension data to avoid incorrect operations by users when the dimension data is generated, and to prevent the amount of dimension data collected from being affected by personal habits or the position of advertisement push, which may cause errors in the user demand assessment and affect the accuracy of the assessment results; According to the analysis of the generation time of the dimension data, the generation methods are divided into active generation and passive generation; active generation and passive generation respectively refer to active search by users or passive clicks on advertisements; passive clicks on advertisements mean that the user clicks too quickly and clicks on the wrong position; The number of active generation times and passive generation times of the user are collected. If the dimension data of the corresponding type of active generation times continuously generates data without interruption during the data collection stage, the corresponding dimension data is marked as continuously concerned data; if the dimension data of the corresponding type of active generation times is generated in a concentrated period during the data collection stage and is not continuously generated, the corresponding dimension data is marked as briefly concerned data; If the ratio of the interval duration between consecutive generation times of the dimension data of the corresponding type of passive generation times to the duration of continuous generation of the dimension data is lower than the duration ratio threshold, the corresponding dimension data is marked as data concerned due to push; if the ratio of the interval duration between consecutive generation times of the dimension data of the corresponding type of passive generation times to the duration of continuous generation of the dimension data is not lower than the duration ratio threshold, the corresponding dimension data is marked as data pushed but not concerned; Each type of dimension data and the corresponding user are uploaded to the service demand assessment platform for storage; The demand prediction lag analysis unit generates a portrait update signal and sends it to the portrait update module; After receiving the portrait update signal, the portrait update module sets portrait tags for the collected user dimension data, performs feature vector transformation according to the portrait tags, and updates the portrait tags in real time through portrait tag setting and corresponding feature vector analysis to facilitate user demand assessment based on the real-time latest portrait tags, improving the accuracy of user demand assessment, avoiding excessive data volume of dimension data corresponding to multiple portrait tags, generating data processing workload for service demand assessment, and prone to errors. By obtaining the latest portrait tags through portrait update, the availability and analysis accuracy of the existing collected data are maximally ensured; Portrait tags are set for each type of dimension data, and the portrait tags are divided into risk preference tags, category preference tags, and demand preference tags; risk preference tags represent risk parameters such as the floating frequency of the usage period of each service demand and the user's demand usage period; category preference tags represent the commodity types of service demands; demand preference tags represent the real-time demand of users for service demands; Convert each type of label into a feature vector, and the feature vector conversion is performed by selecting one within the range of each type of label. For example, the risk deviation label contains multiple data such as the service life and usage cost. Any one of them can be selected for analysis and all data are applicable to this system; Among them, the feature vector of the risk preference label is the excess amount of the service life set for the service demand below the floating frequency threshold compared to the user's required service life; the category preference label is represented by the overlapping browsing frequency of the service demand types in the dimensional data; the demand preference label is represented by the continuous browsing frequency of the user within the service life set for obtaining the service demand; After completing the feature vector conversion, construct a portrait label system with the corresponding data of each type of portrait label, and obtain the ratio of the feature vector value of the corresponding same-type label to the set threshold of the corresponding feature vector according to the service demand selected by the user's history, and mark it as the portrait evaluation ratio; compare the portrait evaluation ratio of each type of real-time label with the portrait evaluation ratio of the historical same-type label in real time. Whether the portrait evaluation ratio of each type of real-time label exceeds the portrait evaluation ratio corresponding to the historical selected service demand is used as the weight setting standard for the current portrait label, that is, set the three portrait labels as high weight, medium weight, and low weight respectively, and construct a portrait label system according to the weights; And send the portrait label system with the weights set to the service demand evaluation platform.

[0023] Embodiment III According to Figure 3 As shown, the service demand evaluation platform is communicatively connected to a service demand evaluation unit; After receiving the portrait label system, the service demand evaluation platform generates a service demand evaluation signal and sends it to the service demand evaluation unit; After receiving the service demand evaluation signal, the service demand evaluation unit conducts an intelligent evaluation of the user's service demand according to the portrait label system, and conducts a service demand evaluation according to the real-time updated portrait label system, effectively improving the accuracy of user evaluation and ensuring the availability of real-time collected data; According to the portrait label system, conduct data analysis on each weighted portrait label in the system, collect the peak increase span of the feature vector value corresponding to the high-weight portrait label and the valley decrease span of the feature vector corresponding to the low-weight portrait label in the portrait label system, and calculate the span value ratio according to the ratio. At the same time, in the floating stage of the feature vector corresponding to the medium-weight portrait label in the portrait label system, obtain the frequency value ratio of the range increased to the high-weight feature vector value and the range decreased to the weight feature vector value; If the span value ratio exceeds the span value ratio threshold, it is inferred that the two-pole weight division of the current portrait label system fits, and the two-pole weights of the current portrait label system are used as the user service demand evaluation criteria, where the two-pole weights are represented as high weight and low weight; If the ratio of span values does not exceed the span value ratio threshold, it is inferred that the weight division between the two poles of the current portrait label system does not fit, and the floating trend of the weights of the two poles corresponding to the current portrait label system is used as the evaluation criterion for user service needs; If the ratio of frequency values exceeds the maximum value of the frequency value ratio range, it is inferred that there is an increase in the data of the portrait labels with high weights in the current portrait label system, and data analysis is performed based on the increased portrait labels to evaluate service needs; If the ratio of frequency values does not exceed the minimum value of the frequency value ratio range, it is inferred that there is an increase in the data of the portrait labels with low weights in the current portrait label system, and data analysis is performed based on the increased portrait labels to evaluate service needs; If the ratio of frequency values is within the frequency value ratio range, it is inferred that the influence of the weight data fluctuation of the current portrait label system is small, and the current portrait label system is used as the evaluation criterion for user service needs; Among them, after determining the user service needs evaluation criterion, user evaluation is carried out. According to the weights of each portrait label in the portrait label system and in combination with the influence of the corresponding dimension data, the service needs of the current user are inferred.

[0024] When the present invention is in use, the data dimension detection unit performs dimension analysis on the data collection in the process of evaluating user service needs; the demand prediction lag analysis unit performs demand prediction lag analysis on the user dimension data collection; after the data dimension detection unit determines that the dimension detection meets the actual requirements, the data screening module screens the collected dimension data and divides each dimension data; the portrait update module sets portrait labels for the collected user dimension data, sets weights for the portrait labels, and constructs a portrait label system according to the portrait labels with different weights; the service needs evaluation unit intelligently evaluates the service needs of the user according to the portrait label system.

[0025] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific embodiments. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can understand and utilize the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. An intelligent evaluation system for user service requirements based on image construction, characterized in that, It includes a service demand assessment platform, and the service demand platform includes: A data dimension detection unit that conducts dimensional analysis on data collection during the user service demand assessment process; A demand prediction lag analysis unit that conducts demand prediction lag analysis on user dimensional data collection; After the data dimension detection unit determines that the dimension detection meets the actual requirements, the data screening module screens the collected dimensional data and divides each dimensional data; The portrait update module sets portrait labels for the collected user dimensional data, sets weights for the portrait labels, and constructs a portrait label system based on the portrait labels with different weights; A service demand assessment unit that conducts intelligent service demand assessment on users according to the portrait label system.

2. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 1, characterized in that The operation process of the data dimension detection unit is as follows: Collect user service demand data, where the service demand is a commodity purchase demand or a consulting service demand; conduct dimension detection on the currently collected data, where the data dimensions are divided into user dimensions, behavior dimensions, and emotion dimensions; when facing different service demands, if there are different numerical manifestations in any dimension data of the user, it is determined that the currently collected data dimensions meet the actual assessment requirements; If there are no different numerical manifestations in any dimension data of the user, conduct single-dimension assessment; When the user generates dimension data, determine the static and dynamic collection of the user data. Statistically analyze the ratio of the number of static collections to the number of dynamic collections during the current collection of each dimension data. If the ratio exceeds the set ratio threshold, or the ratio continues to increase, it is inferred that as the user's dimension data increases, the current data dimension is single and the satisfaction of the dimension data detection requirements shows a downward trend; if the ratio does not exceed the set ratio threshold and does not continue to increase, it is inferred that as the user's dimension data increases, the current data dimension meets the user demand assessment.

3. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 1, characterized in that, The operation process of the demand prediction lag analysis unit is as follows: Divide the user dimension data collection stage into several sub-stages, collect the floating trends of the dimension data in each sub-stage and the floating trend of the dimension data in the entire collection stage. If the corresponding floating trends are inconsistent, mark the sub-stage as a floating stage; if the corresponding floating trends are consistent, mark the sub-stage as a stable stage; During the floating stage, collect the time interval between the moment of trend analysis of the user dimension data and the end moment of the floating stage; During the stable stage, collect the numerical increase span of the corresponding dimension data in multiple stable stages during the trend analysis of the user dimension data; When the time interval during the floating stage exceeds the interval threshold, or the numerical increase span during the stable stage exceeds the span threshold, it is inferred that the lag analysis during the dimension data collection process is abnormal, and the service demand assessment platform shortens the prediction period of the dimension data collection, that is, conducts floating real-time prediction during the floating stage and sets a critical value for the numerical floating span during the stable stage, and conducts demand prediction in a timely manner when it exceeds; when the time interval during the floating stage does not exceed the interval threshold and the numerical increase span during the stable stage does not exceed the span threshold, it is inferred that the lag analysis during the dimension data collection process is normal.

4. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 1, wherein The operation process of the data screening module is as follows: According to the generation mode division based on the moment analysis of dimension data, it is divided into active generation and passive generation; Collect the number of active generations and passive generations of users. If the dimension data of the type corresponding to the number of active generations continuously generates data without interruption during the data collection stage, mark the corresponding dimension data as continuously concerned data; if the dimension data of the type corresponding to the number of active generations is generated in a concentrated period during the data collection stage and is not continuously generated, mark the corresponding dimension data as briefly concerned data; If the ratio of the interval duration between consecutive generation times of the dimension data of the type corresponding to the number of passive generations to the duration of continuous generation of the dimension data is lower than the duration ratio threshold, mark the corresponding dimension data as data concerned due to push; if the ratio of the interval duration between consecutive generation times of the dimension data of the type corresponding to the number of passive generations to the duration of continuous generation of the dimension data is not lower than the duration ratio threshold, mark the corresponding dimension data as pushed but not concerned data.

5. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 1, wherein The operation process of the portrait update module is as follows: Set portrait labels for each type of dimension data, and the portrait labels are divided into risk preference labels, category preference labels, and demand preference labels; perform feature vector transformation on each type of label, and the feature vector transformation is to select one within the range of each type of label for feature vector transformation; Among them, the feature vector of the risk preference label is the excess amount of the service demand with a service demand lower than the floating frequency threshold and the user's demand usage period; the category preference label represents the overlapping browsing frequency of the service demand types in the dimension data; the demand preference label represents the continuous browsing frequency of the user within the service demand usage period set for obtaining the service demand.

6. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 5, characterized in that, After completing the feature vector transformation, construct a portrait label system for the data corresponding to each type of portrait label, and obtain the ratio of the feature vector numerical value of the corresponding same-type label to the set threshold of the corresponding feature vector according to the service demand historically selected by the user, and mark it as the portrait evaluation ratio; compare the portrait evaluation ratio of each type of label in real time with the portrait evaluation ratio of the same type of label in history. Whether the portrait evaluation ratio of each type of label in real time exceeds the portrait evaluation ratio corresponding to the historically selected service demand is used as the weight setting standard for the current portrait label, that is, set the three portrait labels as high weight, medium weight, and low weight respectively, and construct a portrait label system according to the weights.

7. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 1, characterized in that The operation process of the service demand evaluation unit is as follows: According to the portrait label system, perform data analysis on each weighted portrait label in the system, collect the peak increase span of the feature vector numerical value corresponding to the high-weight portrait label and the valley decrease span of the feature vector corresponding to the low-weight portrait label in the portrait label system, and calculate the span numerical ratio according to the ratio. At the same time, in the floating stage of the feature vector corresponding to the medium-weight portrait label in the portrait label system, obtain the frequency numerical ratio of the range increased to the high-weight feature vector numerical value and the range decreased to the weight feature vector numerical value.

8. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 7, characterized in that If the span value ratio exceeds the span value ratio threshold, it is inferred that the two-pole weight division of the current portrait label system is appropriate, and the two-pole weights of the current portrait label system are used as the evaluation criteria for user service needs; if the span value ratio does not exceed the span value ratio threshold, it is inferred that the two-pole weight division of the current portrait label system is inappropriate, and the floating trend of the corresponding two-pole weights of the current portrait label system is used as the evaluation criteria for user service needs; If the frequency value ratio exceeds the maximum value of the frequency value ratio range, it is inferred that there is an increase in the data of the high-weight portrait labels in the current portrait label system, and data analysis is performed based on the increased portrait labels to evaluate service needs; if the frequency value ratio does not exceed the minimum value of the frequency value ratio range, it is inferred that there is an increase in the data of the low-weight portrait labels in the current portrait label system, and data analysis is performed based on the increased portrait labels to evaluate service needs; If the frequency value ratio is within the frequency value ratio range, it is inferred that the influence of the weight data fluctuation of the current portrait label system is small, and the current portrait label system is used as the evaluation criteria for user service needs.

9. The intelligent evaluation system for user service requirements constructed based on portraits according to claim 8, wherein After determining the evaluation criteria for user service needs, user evaluation is carried out. Based on the weights of each portrait label in the portrait label system and combined with the influence of the corresponding dimension data, the service needs of the current user are inferred.

Images