CN107423380B - Information system design method based on user behavior pattern - Google Patents
Information system design method based on user behavior pattern Download PDFInfo
- Publication number
- CN107423380B CN107423380B CN201710569292.4A CN201710569292A CN107423380B CN 107423380 B CN107423380 B CN 107423380B CN 201710569292 A CN201710569292 A CN 201710569292A CN 107423380 B CN107423380 B CN 107423380B
- Authority
- CN
- China
- Prior art keywords
- data
- user
- behavior
- information
- information system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an information system design method based on a user behavior pattern, which comprises the following steps: acquiring behavior mode data, classifying and marking to obtain classified marking data, predicting user behavior, triggering and activating a forgetting function, preparing data information, acquiring user feedback and correcting and activating the forgetting function.
Description
Technical Field
The invention relates to the technical field of information system design, in particular to an information system design method based on a user behavior pattern.
Background
With the progressive development of the information system design theory, the design method of the information system is gradually improved. The mainstream design methods at present include two kinds, one is a top-down structuring method, and some design concepts absorbed by the structuring method in other methods improve the efficiency and quality of system design, for example, a prototype method and an object-oriented method can be used in a local link, and a Use Case (Use Case) is adopted to acquire and understand the functional requirements of the system, and the like. The other mainstream design method is object-oriented design, and the main flow of the method is to make clear and fine the business requirements obtained by analysis, select an effective design style to optimize an object structure, design a system interactive interface, design a database structure and the like. It emphasizes the perfection and improvement of the analysis results, resulting in a detailed specification, i.e. "how to do" description, that guides object-oriented programming.
From the current mainstream design method of the information system, no matter the design of the use case or the object-oriented design, the information system gives corresponding feedback (Response) on the basis that a User (User) of the information system generates a certain behavior (Action), and the feedback may be a page, a visual display, an application system interface or information interaction and the like. The traditional design method cannot predict the next behavior of the user through the operation record information of the information system user, so that the data is prepared in advance.
Disclosure of Invention
Based on the analysis of the background art, the invention aims to solve the technical problem that the existing information system can only react based on the instruction behaviors of users (generally refer to people or systems interacting with the system) and cannot actively predict the user behaviors.
The technical scheme adopted by the invention is as follows:
an information system design method based on user behavior patterns comprises the following steps:
s101: acquiring behavior pattern data; wherein the behavior pattern data is historical behavior pattern data or current behavior pattern data; if the current behavior mode data is the current behavior mode data, the flow is ended after the current behavior mode data is stored; if the historical behavior pattern data is obtained, executing step S102;
s102: according to the service class data which are classified and marked, classifying and marking the historical behavior pattern data to obtain classified and marked data;
s103: predicting the user behavior of the next time node by using a prediction algorithm in machine learning according to the behavior pattern data and the classification marking data, and obtaining a prediction result corresponding to the user behavior; the prediction algorithm is the existing general random forest, support vector machine and the like.
S104: judging whether the prediction result triggers an activation forgetting function or not, and if so, entering the step S105; if not, ending the process;
s105: preparing data information corresponding to the predicted user behavior;
s106: and acquiring user feedback and dynamically correcting an activation forgetting function of the information system according to the user feedback.
In the information system design method based on the user behavior pattern, the behavior pattern data includes four dimensions: time, place, person, event.
In the information system design method based on the user behavior pattern, the current behavior pattern data: when a first user actively initiates a request, detecting that neither the first user nor a second user has a historical access record, and abstracting current behavior mode information of the first user to obtain data; the second user queries the user most similar to the first user by using a similarity evaluation method;
historical behavioral pattern data: when a first user actively initiates a request or a system automatically initiates prediction, it is detected that the first user or a second user has a historical access record, and historical behavior mode data corresponding to the historical access record.
In the information system design method based on the user behavior pattern, the similarity evaluation method adopts the following data: base data and behavioral pattern data; basic data: data registered at system initialization or user registration. The similarity calculation method is based on data of two aspects of users: the system comprises basic data and behavior pattern data, wherein the basic data refers to data registered during system initialization or user registration, and the behavior pattern data is data for measuring behavior habits and interests of users. Assuming that the user similarity calculation function is usersimetric (user a, user B) ═ 0,1, the closer the function value is to 1, indicating that the two users are more similar, here, the specific calculation process of usersimetric may be simple logical comparison, or may be based on time series, deep learning, and the like.
In the method for designing an information system based on a user behavior pattern, the service class data marked by classification specifically includes: service class data which is classified and marked based on human experience and/or by utilizing a classification algorithm in machine learning; the classification algorithm is an existing general algorithm.
The classified service class data has classes and class codes corresponding to the classes one by one, and the class codes have global uniqueness and identifiability.
In the above information system design method based on user behavior patterns, the classification marking based on human experience and/or using a classification algorithm in machine learning occurs before the user does not have human-computer interaction with the information system, and is updated at any time.
In the information system design method based on the user behavior pattern, the prediction result is as follows: predicted user behavior, probability of use of the node at the next time and probability of data access.
In the method for designing an information system based on a user behavior pattern, a preparation mechanism of data information comprises the following steps:
data storage and delayed destruction mechanism: after user behavior prediction is carried out, prepared data information is stored in a memory, and an information system can quickly and accurately access the data information within a limited time range; when the time is out, the information system triggers a data destruction mechanism to remove corresponding data information;
data distribution and sharing mechanisms: and after the user behavior prediction is carried out, storing the prepared data information in a memory, and pushing the data information to the front end of the information system when the user login is detected.
In the information system design method based on the user behavior pattern, the step of obtaining the user feedback and dynamically correcting the activation forgetting function of the information system according to the user feedback specifically comprises the following steps: if the user generates the use behavior or the data access behavior of the prepared data information within the specified time, the forgetting function is activated in a strengthened way; and if the user does not generate the use behavior or the data access behavior of the prepared data information within the specified time, weakening and activating the forgetting function.
In the information system design method based on the user behavior pattern, the system load condition is monitored, and the steps S101-S106 are operated in a cycle mode in a time period with a small system load.
In conclusion, due to the adoption of the technical scheme, the invention has the beneficial effects that:
firstly, the potential use possibility and the data access possibility of the information system are predicted by analyzing the historical operation record information of a user or similar users, data processing and preparation are intelligently carried out at the rear end of the information system, the technical problem that the existing information system can only react based on the instruction behaviors of the user (generally refers to a person or a system interacting with the system) and cannot actively predict the user behaviors is solved, and the method has extremely high innovation, intelligence and convenience.
The method and the system have the advantages that the possibility of the occurrence of the user behavior is predicted in advance, and the data information is prepared in advance, so that the system working time during the user access is saved, and the access efficiency is greatly improved.
Thirdly, the possibility of the occurrence of the user behavior is predicted when the load is small, and data information is prepared in advance, so that the heavy load of the cpu and the memory is greatly reduced; meanwhile, when the load is small, the execution avoids influencing the operation of other programs.
And fourthly, designing the information system, namely establishing a bottom layer design mechanism of the information system, and after the data preparation is finished, making visible informatization expression and an invisible information processing mechanism to meet various use requirements and experiences of customers.
Drawings
FIG. 1 is a framework flow diagram of the present invention;
FIG. 2 is a flowchart of example 1.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The present invention will be described in detail with reference to fig. 1 and 2.
As shown in fig. 1, an information system design method based on user behavior patterns includes the following steps:
s101: acquiring behavior pattern data; wherein the behavior pattern data is historical behavior pattern data or current behavior pattern data; if the current behavior mode data is the current behavior mode data, the flow is ended after the current behavior mode data is stored; if the historical behavior pattern data is obtained, executing step S102;
s102: according to the service class data which are classified and marked, classifying and marking the historical behavior pattern data to obtain classified and marked data;
s103: predicting the user behavior of the next time node by using a prediction algorithm in machine learning according to the behavior pattern data and the classification marking data, and obtaining a prediction result corresponding to the user behavior;
s104: judging whether the prediction result triggers an activation forgetting function or not, and if so, entering the step S105; if not, ending the process;
s105: preparing data information corresponding to the predicted user behavior;
s106: and acquiring user feedback and dynamically correcting an activation forgetting function of the information system according to the user feedback.
In the information system design method based on the user behavior pattern, the behavior pattern data comprises four dimensions: time, place, person, event.
In the information system design method based on the user behavior pattern, the current behavior pattern data: when a first user actively initiates a request, detecting that neither the first user nor a second user has a historical access record, and abstracting current behavior mode information of the first user to obtain data; the second user queries the user most similar to the first user by using a similarity evaluation method;
historical behavioral pattern data: when a first user actively initiates a request or a system automatically initiates prediction, it is detected that the first user or a second user has a historical access record, and historical behavior mode data corresponding to the historical access record.
In the information system design method based on the user behavior pattern, the similarity evaluation method adopts the following data: base data and behavioral pattern data; basic data: data registered at system initialization or user registration. The similarity calculation method is based on data of two aspects of users: the system comprises basic data and behavior pattern data, wherein the basic data refers to data registered during system initialization or user registration, and the behavior pattern data is data for measuring behavior habits and interests of users. Assuming that the user similarity calculation function is usersimetric (user a, user B) ═ 0,1, the closer the function value is to 1, indicating that the two users are more similar, here, the specific calculation process of usersimetric may be simple logical comparison, or may be based on time series, deep learning, and the like.
In the information system design method based on the user behavior pattern, the classified service class data are specifically as follows: service class data which is classified and marked based on human experience and/or by utilizing a classification algorithm in machine learning; the classification marking is carried out based on human experience and/or by utilizing a classification algorithm in machine learning, and occurs before a user does not interact with the information system in a man-machine interaction mode, and is updated at any time. The classified service class data has classes and class codes corresponding to the classes one by one, and the class codes have global uniqueness and identification.
In the information system design method based on the user behavior pattern, the prediction result is as follows: predicted user behavior, probability of use of the node at the next time and probability of data access.
In the information system design method based on the user behavior pattern, the preparation mechanism of the data information comprises the following steps: data storage and delayed destruction mechanism: after user behavior prediction is carried out, prepared data information is stored in a memory, and an information system can quickly and accurately access the data information within a limited time range; when the time is out, the information system triggers a data destruction mechanism to remove corresponding data information; data distribution and sharing mechanisms: and after the user behavior prediction is carried out, storing the prepared data information in a memory, and pushing the data information to the front end of the information system when the user login is detected.
In the information system design method based on the user behavior mode, the specific steps of acquiring the user feedback and dynamically correcting the activation forgetting function of the information system according to the user feedback are as follows: if the user generates a using behavior or a data access behavior of the prepared data information within a specified time, the forgetting function is activated in a strengthened manner; and if the user does not generate the use behavior or the data access behavior of the prepared data information within the specified time, weakening the activation forgetting function.
In the information system design method based on the user behavior pattern, the system load condition is monitored, and the steps S101-S106 are operated circularly in a time period with a small system load.
Example 1
As shown in fig. 2:
(1) and a user A (generally a person or a system interacting with the system) uses the enterprise economic benefit analysis function of the system in 2017, 5 and 22, if the system detects that the user A or a user most similar to the user A has no historical access record before 2017, 5 and 22, the current behavior pattern data are obtained and stored, otherwise, historical behavior pattern data (a preset value can be adopted) of the information system with the maximum similarity are found by using a similarity evaluation method. In this example, historical behavior pattern data (system presets of 20170518 and 20170522 within 5 days before the occurrence of the current behavior) is employed.
(2) And classifying and marking the acquired behavior pattern data, inquiring the category of the behavior pattern data according to the classified and marked service category data table 2, and classifying and marking the behavior pattern data according to category codes corresponding to the categories to obtain classified and marked data. As shown in Table 1, Table 1 shows some types of user behavior pattern data abstracted by the system and classified labeled data C1/C2, and Table 2 shows traffic class data classified and labeled by using classification algorithms available in human experience or machine learning.
TABLE 1
| User ID time | Time of day | Other Properties | Event(s) | Data class ID |
| 001 | 20170518 | … | Look up workshop weekly production report | C1 |
| 001 | 20170519 | … | Consulting workshop monthly production report | C1 |
| 002 | 20170519 | … | Look up the yield of production orders | C2 |
| 001 | 20170520 | … | Look up workshop weekly production report | C1 |
| 001 | 20170521 | … | Look up workshop weekly production report | C1 |
TABLE 2
| Data class ID | Class of service | Field set |
| C1 | In-process of production | Workshop, product line, quantity of work in process |
| C2 | Quality of product | Material coding, product coding and semi-finished product |
(3) Assuming that the activation and forgetting function of the system is f (x), when f (x) > 75%, a trigger signal is output to step (4) (in an actual application system, f (x) may be a very complex combination function). In this example, let f (x) denote the frequency of accessing the same category data within 5 days of the same user, that is, f (x) is 80%, and step (4) receives the trigger signal.
(4) The trigger signal triggers a specific data extraction operation. And C1-type related data information is extracted specifically.
(5) And (4) actively pushing the data information generated in the step (4) to the user A by a preparation mechanism of the information system, and recording the feedback of the user A to the data within a specified time.
(6) According to the feedback of the user A to the push data received by the system, the system perception function is realized, namely the activation and forgetting functions of the system are dynamically corrected: if the system determines that the data pushed by the user a is valid, the activation function is rewarded, for example, f (x) ═ f (x) + 5%, otherwise, the activation function is punished, and f (x) ═ f (x) -5%.
(7) And (4) according to the availability plan of the system load, circulating the steps (1) - (6) by using a time period with smaller system load so as to improve the efficiency and the intelligent degree of the system.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An information system design method based on user behavior patterns is characterized by comprising the following steps:
s101: acquiring behavior pattern data; wherein the behavior pattern data is historical behavior pattern data or current behavior pattern data; if the current behavior mode data is the current behavior mode data, the flow is ended after the current behavior mode data is stored; if the historical behavior pattern data is obtained, executing step S102;
s102: according to the service class data which are classified and marked, classifying and marking the historical behavior pattern data to obtain classified and marked data;
s103: predicting the user behavior of the next time node by using a prediction algorithm in machine learning according to the historical behavior pattern data and the classification marking data, and obtaining a prediction result corresponding to the user behavior;
s104: judging whether the prediction result triggers an activation forgetting function or not, and if so, entering the step S105; if not, ending the process;
s105: preparing data information corresponding to the predicted user behavior;
s106: and acquiring user feedback and dynamically correcting an activation forgetting function of the information system according to the user feedback.
2. The method of claim 1, wherein the behavior pattern data comprises four dimensions: time, place, person, event.
3. The method of claim 1, wherein the information system design method based on user behavior patterns,
the current behavior pattern data: when a first user actively initiates a request, detecting that neither the first user nor a second user has a historical access record, and abstracting current behavior mode information of the first user to obtain data; the second user queries the user most similar to the first user by using a similarity evaluation method;
historical behavioral pattern data: when a first user actively initiates a request or a system automatically initiates prediction, it is detected that the first user or a second user has a historical access record, and historical behavior mode data corresponding to the historical access record.
4. The method of claim 3, wherein the similarity evaluation method uses data comprising: base data and behavioral pattern data;
basic data: data registered at system initialization or user registration.
5. The method according to claim 1, wherein the service class data labeled by classification specifically comprises: service class data which is classified and marked based on human experience and/or by utilizing a classification algorithm in machine learning;
the classified service class data has classes and class codes corresponding to the classes one by one, and the class codes have global uniqueness and identifiability.
6. The method as claimed in claim 5, wherein the classification based on human experience and/or classification algorithm in machine learning is performed before the user does not interact with the information system and is updated at any time.
7. The method of claim 1, wherein the prediction result is: predicted user behavior, probability of use of the node at the next time and probability of data access.
8. The method of claim 1, wherein the mechanism for preparing data information comprises:
data storage and delayed destruction mechanism: after user behavior prediction is carried out, prepared data information is stored in a memory, and an information system can quickly and accurately access the data information within a limited time range; when the time is out, the information system triggers a data destruction mechanism to remove corresponding data information;
data distribution and sharing mechanisms: and after the user behavior prediction is carried out, storing the prepared data information in a memory, and pushing the data information to the front end of the information system when the user login is detected.
9. The information system design method based on the user behavior pattern as claimed in claim 1, wherein the obtaining of the user feedback and the dynamic modification of the activation forgetting function of the information system according to the user feedback are specifically: if the user generates the use behavior or the data access behavior of the prepared data information within the specified time, the forgetting function is activated in a strengthened way; and if the user does not generate the use behavior or the data access behavior of the prepared data information within the specified time, weakening and activating the forgetting function.
10. The method according to claim 1, wherein the system load is monitored, and the steps S101-S106 are cyclically executed in a period of time when the system load is small.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710569292.4A CN107423380B (en) | 2017-07-13 | 2017-07-13 | Information system design method based on user behavior pattern |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710569292.4A CN107423380B (en) | 2017-07-13 | 2017-07-13 | Information system design method based on user behavior pattern |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107423380A CN107423380A (en) | 2017-12-01 |
| CN107423380B true CN107423380B (en) | 2020-09-11 |
Family
ID=60427491
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710569292.4A Active CN107423380B (en) | 2017-07-13 | 2017-07-13 | Information system design method based on user behavior pattern |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107423380B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101189632A (en) * | 2005-05-02 | 2008-05-28 | 瑞士联合银行集团 | Data processing method for time optimal computation of large result data sets |
| CN101247620A (en) * | 2008-03-24 | 2008-08-20 | 中兴通讯股份有限公司 | Method for realizing user-defined report forms |
| CN102111920A (en) * | 2009-12-23 | 2011-06-29 | 大唐移动通信设备有限公司 | Method and device for managing performance report |
| CN105807741A (en) * | 2016-03-09 | 2016-07-27 | 北京科技大学 | Industrial production flow prediction method |
| CN106200434A (en) * | 2016-08-18 | 2016-12-07 | 广州市迪士普音响科技有限公司 | The cloud computing of wisdom platform and big data processing method and system |
| CN106777037A (en) * | 2016-12-09 | 2017-05-31 | 南京欣网互联信息技术有限公司 | A kind of method for generating operation platform integration form |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140272849A1 (en) * | 2013-03-15 | 2014-09-18 | Yahoo! Inc. | System and method providing positive social and economic motivators for goal achievement |
-
2017
- 2017-07-13 CN CN201710569292.4A patent/CN107423380B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101189632A (en) * | 2005-05-02 | 2008-05-28 | 瑞士联合银行集团 | Data processing method for time optimal computation of large result data sets |
| CN101247620A (en) * | 2008-03-24 | 2008-08-20 | 中兴通讯股份有限公司 | Method for realizing user-defined report forms |
| CN102111920A (en) * | 2009-12-23 | 2011-06-29 | 大唐移动通信设备有限公司 | Method and device for managing performance report |
| CN105807741A (en) * | 2016-03-09 | 2016-07-27 | 北京科技大学 | Industrial production flow prediction method |
| CN106200434A (en) * | 2016-08-18 | 2016-12-07 | 广州市迪士普音响科技有限公司 | The cloud computing of wisdom platform and big data processing method and system |
| CN106777037A (en) * | 2016-12-09 | 2017-05-31 | 南京欣网互联信息技术有限公司 | A kind of method for generating operation platform integration form |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107423380A (en) | 2017-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10484413B2 (en) | System and a method for detecting anomalous activities in a blockchain network | |
| Yan et al. | On machine learning towards predictive sales pipeline analytics | |
| CN105590055B (en) | Method and device for identifying user credible behaviors in network interaction system | |
| Köhn et al. | Hierarchical cluster analysis | |
| US10504120B2 (en) | Determining a temporary transaction limit | |
| US20170024446A1 (en) | Data storage extract, transform and load operations for entity and time-based record generation | |
| US20140351285A1 (en) | Platform and method for analyzing electric power system data | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| Patibandla et al. | Survey on clustering algorithms for unstructured data | |
| US11443234B2 (en) | Machine learning data processing pipeline | |
| Hasugian | The data mining of cell phone most interested using apriorial algorithm | |
| CN106294406B (en) | Method and equipment for processing application access data | |
| CN105654277A (en) | Transaction operation identification method and server | |
| CN115879017A (en) | Automatic classification and grading method and device for power sensitive data and storage medium | |
| CN113569910A (en) | Account type identification method and device, computer equipment and storage medium | |
| CN107423380B (en) | Information system design method based on user behavior pattern | |
| Ganghishetti et al. | Association rule mining via evolutionary multi-objective optimization | |
| CN113392399A (en) | Malicious software classification method, device, equipment and medium | |
| CN115982646B (en) | Management method and system for multisource test data based on cloud platform | |
| CN109542953B (en) | Presto-based data processing method and device | |
| Lee et al. | Detecting anomaly teletraffic using stochastic self-similarity based on Hadoop | |
| CN113918577A (en) | Data table identification method and device, electronic equipment and storage medium | |
| CN113127465A (en) | Data fusion method and system | |
| TWM578844U (en) | Establishing user tag library and marketing system by applying artificial intelligence to insurance service | |
| Shen et al. | Customer churn prediction model for telecommunication industry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |