WO2020010570A1 - Big data automatic analysis system based on consumer habits - Google Patents

Abstract

A big data automatic analysis system based on consumer habits, the system comprising a server and a plurality of mobile terminals connected to the server; the server executes the following operations: a data collection module collects habit data of a mobile terminal user and stores said data in real time when the server interacts with a mobile terminal; a data analyzing module analyzes the habit data of the user and forms an associated knowledge set, the associated knowledge set being formed by: classifying the user habit data and deep mining each category to form knowledge sets, and then performing association calculation on each knowledge set to form the associated knowledge set; a request receiving module obtains a request sent by the mobile terminal; a determining module determines, according to the request received by the request receiving module, the knowledge set formed by the data analyzing module as well as a preset threshold condition, and forms a determination result; a feedback module feeds back the result of the determining module to the mobile terminal. By means of the described manner, more accurate personalized services may be provided to customers.

Description

Big data automatic analysis system based on consumer habits Technical field

The invention relates to a method for implementing big data analysis based on consumer habits.

Background technique

With the rapid development of science and technology, society and economy, the era of big data has begun. The value source of big data has more comprehensive characteristics. From the data dimension, time dimension, spatial dimension, and cross-border data, they all converge. The value of big data is reflected in more accurate and personalized services for customers.In addition, big data has great value potential in artificial intelligence such as language, vision, and prediction.For example, behavior analysis based on consumer habits can give Users bring many significant conveniences, which can provide customers with precise and extreme personalized services.

On the other hand, with the continuous progress of society, the degree of a commoditized society has continued to increase, and the consumption power of residents has continued to increase, while the continuous progress of the information society has led to the diversity and multi-selection of residents' consumption patterns, which poses a challenge to the commerce industry Will be more intense. How to judge the consumption habits of end consumers, and thus formulate their own product sales strategies has become a top priority.

Summary of the invention

A brief overview of embodiments of the invention is given below in order to provide a basic understanding of certain aspects of the invention. It should be understood that the following summary is not an exhaustive overview of the invention. It is not intended to identify key or important parts of the invention, nor is it intended to limit the scope of the invention. Its purpose is merely to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later.

According to one aspect of the present application, an automatic big data analysis system based on consumer habits is provided. The system includes a server and a plurality of mobile terminals connected to the server. The server is provided with at least: a user for collecting mobile terminals. A data collection module for habit data, a data analysis module for analyzing user habit data, a reception request module for receiving a request from a mobile terminal, and a judgment module for determining a request from a mobile terminal based on the data analysis module. To feedback the result of the judgment module to the feedback module of the mobile terminal; the server performs the following operations:

The data collection module collects and stores mobile terminal user habits data in real time when the server interacts with the mobile terminal;

The data analysis module analyzes user habit data and forms a related knowledge set. The related knowledge set is: classify the user habit data and perform deep mining for each category to form a knowledge set, and then perform an association calculation on each knowledge set to form a related knowledge set. ;

The receiving request module acquires a request sent by a mobile terminal; specifically, the receiving request module acquires multiple requests sent by the mobile terminal within a preset time period, and summarizes the multiple requests into an association request set;

The judgment module makes a judgment according to the request received by the reception request module, the knowledge set formed by the data analysis module, and a preset threshold condition, and forms a judgment result;

The feedback module feeds back the result of the judgment module to the mobile terminal.

Further, the data analysis module analyzes user habit data and forms a related knowledge set. The related knowledge set is: classify the user habit data and perform in-depth mining for each category to form a knowledge set, and then calculate each knowledge set through association calculation. Related knowledge sets, including:

The user habit data includes the user subjective attribute set S and the user objective attribute set O; the user subjective attribute set S is recorded as {S1, S2, ..., Sm}, and m is the number of subjective attributes of the user subjective attribute set S, Among them, the attribute characteristic value of Si is recorded as {si1, si2, ..., sit}, 1 <i <m, t is the user's subjective attribute set S is a natural number, which refers to the number of attribute characteristic values; the user's objective attribute set O Denote as {O1, O2, ..., On}, n is the number of objective attributes of the user objective attribute set O, where the attribute characteristic value of Oj is denoted as {oj1, oj2, ..., ojr}, 1 <j <n, r is the number of attribute characteristic values of the user's subjective attribute set O, which is a natural number; wherein the user's subjective attribute set S is used to describe the user's preferences and understand the user's psychological needs; the user's objective attribute set O It refers to the data feedback of the user using the application of the mobile terminal to understand the user's frequently occurring behavior.

Deeply mine each attribute Si of the user's subjective attribute set S, and integrate the attribute characteristic values of Si {si1, si2, ..., sit} according to preset rules to form a knowledge set Si ', Si' = { si1 ', si2', ..., sit '};

Perform deep mining for each attribute Oj of the user's objective attribute set O, and integrate the attribute characteristic values {oj1, oj2, ..., ojr} of Oj according to preset rules to form a knowledge set Oj ', Oj' = { oj1 ', oj2', ..., ojt '};

According to the preset knowledge rules, the correlation calculation is performed on the knowledge set Si 'and the knowledge set Oj' to obtain the related knowledge set R. The related knowledge set R is denoted as {R1, R2, ..., Ra}, and a is the value of the related knowledge set. The number of associated attributes.

The calculation of the correlation between the knowledge set Si ′ and the knowledge set Oj ’according to a preset knowledge rule is specifically:

The knowledge set Si 'and the knowledge set Oj' are classified separately, and the knowledge sets describing the same class are associated. The association algorithm is: preset the subjective weight value of the knowledge set Si 'is δ, δ ∈ [0,1], Then the objective weight value of the knowledge set Oj 'is 1-δ, then the i-type associated knowledge set Ri (1 <i <a) = δΣSi' + (1-δ) ΣOj ', ΣSi' belong to the same category Sum of category i subjective knowledge set, ΣOj 'is the sum of category i objective knowledge set; In addition, the subjective weight value can be customized by the user, and the system sets the initial value of subjective weight value δ to δ0, δ0 ∈ [0,1], if the user defines it, the system setting value will be replaced by the delta δ. If the user does not define it, it will be calculated by the system setting.

Further, the obtaining of the request sent by the mobile terminal by the receiving request module is specifically that the receiving request module obtains multiple requests sent by the mobile terminal within a preset time period, and summarizes the multiple requests into an association request set.

The judgment module makes a judgment according to the request received by the reception request module, the knowledge set formed by the data analysis module, and a preset threshold condition, and forms a judgment result, specifically:

The receiving request module obtains multiple requests sent by the mobile terminal within a preset time period, and summarizes the multiple requests into an association request set A, A = {A1, A2,…, Az}; the attribute characteristic value of Ck is recorded {Ck1, ck2,…, ckz}, 1 <k <z, where z is the number of attribute eigenvalues of Ck;

Classify the request (the classification rules are the same as the classification rules of the associated knowledge set) and get the classification results;

Obtain multiple related knowledge sets in the related knowledge set R that are the same as the requested classification result;

Calculate the matching degree between multiple related knowledge sets and the request, and get the judgment result. Among them, the judgment result may be a related knowledge set with the highest matching degree, or may be multiple related knowledge sets listed in order according to the matching degree from high to low.

Compared with the prior art, the present invention has the following advantages:

1. Creatively divide user habits into subjective and objective aspects, that is, user subjective attribute set S and user objective attribute set O, which can provide accurate personalized services for customer data mining. Among them, user subjective attribute set S is used In describing the user's preferences, the user's psychological needs can be understood; the user's objective attribute set O is used to describe the data feedback of the user's application using the mobile terminal, and can understand the user's frequently occurring behavior;

2. According to the present invention, the subjective knowledge set Si 'and the objective knowledge set Oj' after mining are subjected to correlation calculation to obtain a further related knowledge set. When the judgment module judges a user's request, it is easy to pass the knowledge set Retrieve results quickly and intelligently. Retrieval in the prior art generally uses scattered data for retrieval or evaluation of retrieval formulas to obtain better results. The present invention classifies back-end data as a knowledge set one by one, which can not only speed up Retrieval speed, and can also accurately retrieve the customized information required by users, has a very good human experience.

detailed description

Embodiments of the present invention will be described below.

The invention provides a big data automatic analysis system based on consumer habits. The system includes a server and a plurality of mobile terminals connected to the server. The server is provided with at least a data collection module for collecting user habits data of the mobile terminal. A data analysis module for analyzing user habits data, a reception request module for receiving a request from a mobile terminal, a determination module for determining a request from a mobile terminal based on the data analysis module, and a feedback for a result of the determination module Feedback module for mobile terminal.

The server performs the following operations:

The data collection module collects and stores mobile terminal user habits data in real time when the server interacts with the mobile terminal;

The data analysis module analyzes user habit data and forms a related knowledge set. The related knowledge set is: classify the user habit data and perform deep mining for each category to form a knowledge set, and then perform an association calculation on each knowledge set to form a related knowledge set. ;

The receiving request module acquires a request sent by a mobile terminal; specifically, the receiving request module acquires multiple requests sent by the mobile terminal within a preset time period, and summarizes the multiple requests into an association request set;

The judgment module makes a judgment according to the request received by the reception request module, the knowledge set formed by the data analysis module, and a preset threshold condition, and forms a judgment result;

The feedback module feeds back the result of the judgment module to the mobile terminal.

Among them, the data analysis module analyzes user habits data and forms a related knowledge set. The related knowledge set is: classify the user habits data and perform deep mining for each category to form a knowledge set, and then calculate the association of each knowledge set to form a relationship. Knowledge set, including:

The user habit data includes the user subjective attribute set S and the user objective attribute set O; the user subjective attribute set S is recorded as {S1, S2, ..., Sm}, and m is the number of subjective attributes of the user subjective attribute set S, Among them, the attribute characteristic value of Si is recorded as {si1, si2, ..., sit}, 1 <i <m, t is the user's subjective attribute set S is a natural number, which refers to the number of attribute characteristic values; the user's objective attribute set O Denote as {O1, O2, ..., On}, n is the number of objective attributes of the user's objective attribute set O, where the attribute characteristic value of Oj is recorded as {oj1, oj2, ..., ojr}, 1 <j <n, r is the number of attribute characteristic values of the user's subjective attribute set O, which is a natural number; where the user's subjective attribute set S is used to describe the user's preferences and understand the user's psychological needs; the user's objective attribute set O refers to The user uses the data feedback of the mobile terminal application to understand the user's recurring behavior.

Deeply mine each attribute Si of the user's subjective attribute set S, and integrate the attribute characteristic values of Si {si1, si2, ..., sit} according to preset rules to form a knowledge set Si ', Si' = { si1 ', si2', ..., sit '};

Perform deep mining for each attribute Oj of the user's objective attribute set O, and integrate the attribute characteristic values {oj1, oj2, ..., ojr} of Oj according to preset rules to form a knowledge set Oj ', Oj' = { oj1 ', oj2', ..., ojt '};

According to the preset knowledge rules, the correlation calculation is performed on the knowledge set Si 'and the knowledge set Oj' to obtain the related knowledge set R. The related knowledge set R is denoted as {R1, R2, ..., Ra}, and a is the value of the related knowledge set. The number of associated attributes.

The calculation of the correlation between the knowledge set Si ′ and the knowledge set Oj ’according to a preset knowledge rule is specifically:

The knowledge set Si 'and the knowledge set Oj' are classified separately, and the knowledge sets describing the same class are associated. The association algorithm is: preset the subjective weight value of the knowledge set Si 'is δ, δ ∈ [0,1], Then the objective weight value of the knowledge set Oj 'is 1-δ, then the i-type associated knowledge set Ri (1 <i <a) = δΣSi' + (1-δ) ΣOj ', ΣSi' belong to the same category Sum of category i subjective knowledge set, ΣOj 'is the sum of category i objective knowledge set; In addition, the subjective weight value can be customized by the user, and the system sets the initial value of subjective weight value δ to δ0, δ0 ∈ [0,1], if the user defines it, the system setting value will be replaced by the delta δ. If the user does not define it, it will be calculated by the system setting.

The judgment module makes a judgment according to the request received by the reception request module, the knowledge set formed by the data analysis module, and a preset threshold condition, and forms a judgment result, specifically:

The receiving request module obtains multiple requests sent by the mobile terminal within a preset time period, and summarizes the multiple requests into an association request set A, A = {A1, A2,…, Az}; the attribute characteristic value of Ck is recorded {Ck1, ck2,…, ckz}, 1 <k <z, where z is the number of attribute eigenvalues of Ck;

Classify the request (the classification rules are the same as the classification rules of the associated knowledge set) and get the classification results;

Obtain multiple related knowledge sets in the related knowledge set R that are the same as the requested classification result;

Calculate the matching degree between multiple related knowledge sets and the request, and get the judgment result. Among them, the judgment result may be a related knowledge set with the highest matching degree, or may be multiple related knowledge sets listed in order according to the matching degree from high to low.

The calculation of the matching degree between the related knowledge set and the request is specifically: extracting the first w-th related knowledge set Rw, and matching the characteristic feature values of each attribute of the request Ck (1 <k <z) with Rw to obtain The matching values are {match_ck1, match_ck2, ..., match_ckz}, and the comprehensive matching value match_w1 is the weighted average of each matching value. The second w-type associated knowledge set Rw is extracted and the above operation is repeated to obtain match_w2 ... match_wz. The comprehensive matching values match_w1, match_w2, ... match_wz are arranged according to the numerical value, and the corresponding related knowledge set is arranged in order from high to low as the judgment result.

User habits refer to a user's recurring behavior. Essentially, user habits include the frequency with which the user uses the product, and the user's determination of the utility of the product (perceivedutility). The invention creatively divides the user habits into subjective and objective aspects, that is, the user's subjective attribute set S and the user's objective attribute set O, which can provide accurate personalized services for customer data mining, where subjective refers to the user's preference (Data that reflects the user ’s personality and preferences can be classified as subjective data. For example, when users browse news or articles or books, they often click pessimistic or full of righteousness, that is, subjective data.) Subjective data can understand the user's psychological needs. ; Objective refers to the data feedback of the user's application using the mobile terminal (for example, the user often uses the application store APPStore to download the APP application, then the application store's data feedback is objective data), and the objective data can understand the user's recurring behavior; At the same time, according to the present invention, the subjective knowledge set Si 'and the objective knowledge set Oj' after mining are subjected to correlation calculation to obtain a further related knowledge set. When the judgment module judges a user's request, it is easy to pass the knowledge set. Retrieve results quickly and intelligently. The search in the prior art is generally decentralized data. For example, when a user searches for a "mobile phone", the retrieved result is generally a mobile phone encyclopedia and a mobile-related hot news. In the present invention, a user search for a "mobile phone" is based on the user. Preference directly recommends a certain mobile phone and the related knowledge list of the mobile phone (the mobile phone's sales, praise rate, repair rate and other user data feedback), so as to provide customers with more accurate personalized services.

The invention not only improves the user experience mentioned above, but also improves the intelligent characteristics: when users search for problems, they no longer need to summarize their knowledge points by themselves (the search results in the prior art are one-to-one knowledge points that are not to be controlled. ), But directly gives the related knowledge set after induction, which has very good intelligent characteristics.

It should be emphasized that the term "including / comprising" as used herein refers to the presence of a feature, element, step or component, but does not exclude the presence or addition of one or more other features, elements, steps or components.

In addition, the method of the present invention is not limited to being performed in the chronological order described in the specification, but may also be performed in other chronological order, in parallel, or independently. Therefore, the execution order of the methods described in this specification does not limit the technical scope of the present invention.

Although the present invention has been disclosed above by describing specific embodiments of the present invention, it should be understood that all the embodiments and examples described above are exemplary and not restrictive. Those skilled in the art may design various modifications, improvements, or equivalents to the present invention within the spirit and scope of the appended claims. These modifications, improvements or equivalents should also be considered to be included in the protection scope of the present invention.

Claims (8)

1. An automatic big data analysis system based on consumer habits, which is characterized in that the system includes a server and a plurality of mobile terminals connected to the server;

   The server is provided with at least: a data collection module for collecting user habits data of the mobile terminal, a data analysis module for analyzing user habits data, a receiving request module for receiving requests from the mobile terminal, and A data analysis module, a judgment module that judges a request of a mobile terminal, and is configured to feed back the result of the judgment module to a feedback module of the mobile terminal;

   The server performs the following operations:

   The data collection module collects and stores mobile terminal user habits data in real time when the server interacts with the mobile terminal;

   The data analysis module analyzes user habit data and forms a related knowledge set. The related knowledge set is: classify the user habit data and perform deep mining for each category to form a knowledge set, and then perform an association calculation on each knowledge set to form a related knowledge set. ;

   The receiving request module obtains a request sent by a mobile terminal;

   The judgment module makes a judgment according to the request received by the reception request module, the knowledge set formed by the data analysis module, and a preset threshold condition, and forms a judgment result;

   The feedback module feeds back the result of the judgment module to the mobile terminal.
2. The big data automatic analysis system based on consumer habits according to claim 1, characterized in that:

   The user habit data includes a user subjective attribute set S and a user objective attribute set O;

   The user subjective attribute set S is recorded as {S1, S2, ..., Sm}, and m is the number of subjective attributes of the user subjective attribute set S. Among them, the attribute characteristic value of Si is recorded as {si1, si2, ... , Sit}, 1 <i <m, t is the user's subjective attribute set S is a natural number, which refers to the number of attribute characteristic values;

   The user objective attribute set O is recorded as {O1, O2, ..., On}, and n is the number of objective attributes of the user objective attribute set O. Among them, the attribute characteristic value of Oj is recorded as {oj1, oj2, ... , Ojr}, 1 <j <n, r is the number of attribute characteristic values of the user's subjective attribute set O, and is a natural number.
3. The big data automatic analysis system based on consumer habits according to claim 2, characterized in that:

   The data analysis module analyzes user habits data and forms related knowledge sets, including:

   Deeply mine each attribute Si of the user's subjective attribute set S, and integrate the attribute characteristic values of Si {si1, si2, ..., sit} according to preset rules to form a knowledge set Si ', Si' = { si1 ', si2', ..., sit '};

   Perform deep mining for each attribute Oj of the user's objective attribute set O, and integrate the attribute characteristic values {oj1, oj2, ..., ojr} of Oj according to preset rules to form a knowledge set Oj ', Oj' = { oj1 ', oj2', ..., ojt '};

   According to the preset knowledge rules, the correlation calculation is performed on the knowledge set Si 'and the knowledge set Oj' to obtain the related knowledge set R. The related knowledge set R is denoted as {R1, R2, ..., Ra}, and a is the value of the related knowledge set. The number of associated attributes.
4. The big data automatic analysis system based on consumer habits according to claim 3, characterized in that: the user's subjective attribute set S is used to describe the user's preferences and understand the user's psychological needs; the user's objective attribute set O refers to the user Use data feedback from mobile terminal applications to understand the recurring behaviors of users.
5. The big data automatic analysis system based on consumer habits according to claim 3 or 4, characterized in that the correlation calculation of the knowledge set Si 'and the knowledge set Oj' according to a preset knowledge rule is specifically: 'And the knowledge set Oj' are classified separately, and the knowledge sets describing the same class are associated. The association algorithm is: preset the subjective weight value of the knowledge set Si 'δ, δ ∈ [0,1], then the knowledge set Oj 'The objective weight value is 1-δ, then the i-type related knowledge set Ri (1 <i <a) = δΣSi' + (1-δ) ΣOj ', ∑Si' is also subjective of the i-th category The sum of knowledge sets, ΣOj 'is the sum of objective knowledge sets belonging to the i category.
6. The big data automatic analysis system based on consumer habits according to claim 5, characterized in that: the receiving request module acquires a request sent by the mobile terminal; and specifically, the receiving request module acquires a plurality of data sent by the mobile terminal within a preset time period. Requests and summarize the multiple requests into a set of associated requests.
7. The big data automatic analysis system based on consumer habits according to claim 6, characterized in that the judgment module judges according to the request received by the receiving request module, the knowledge set formed by the data analysis module, and a preset threshold condition, and forms The judgment results are:

   The receiving request module obtains multiple requests sent by the mobile terminal within a preset time period, and summarizes the multiple requests into an association request set A, A = {A1, A2,…, Az}; the attribute characteristic value of Ck is recorded {Ck1, ck2,…, ckz}, 1 <k <z, where z is the number of attribute eigenvalues of Ck;

   Classify the request and get the classification result;

   Obtain multiple related knowledge sets in the related knowledge set R that are the same as the requested classification result;

   Calculate the matching degree between multiple related knowledge sets and the request, and get the judgment result.
8. The big data automatic analysis system based on consumer habits according to claim 7, characterized in that: calculating the matching degree between the associated knowledge set and the request is specifically:

   Extract the first w-th related knowledge set Rw, and match the characteristic feature values of each request Ck (1 <k <z) with Rw to obtain the respective matching values {match_ck1, match_ck2, ..., match_ckz}, Calculate the comprehensive matching value match_w1 as the weighted average of each matching value;

   Extract the second w-type associated knowledge set Rw and repeat the above operation to obtain match_w2 ... match_wz;

   The comprehensive matching values match_w1, match_w2, ... match_wz are arranged according to the numerical value.