Prevarum behavior identification method, system and equipment and storage medium
Through multi-dimensional outlier detection of task single data and natural language processing technology identification of shirking behavior, the problem of low manual audit efficiency in the existing technology is solved, and more efficient and accurate shirking behavior identification is achieved.
Patent Information
- Application Number
- CN202411186377.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the existing technology, the identification of shirking behavior depends on manual review, which is large in workload, low in efficiency and strong subjectivity, resulting in low judgment accuracy.
By obtaining task order data, calculating outliers of processing time and number of processing times, and using natural language processing models for sentiment analysis, combining flow path identification with irregular order refunds, comprehensively determine whether there is any shirkish behavior.
The work efficiency and accuracy of identifying shirk behaviors is improved, and the shirk behaviors in the task order data are fully covered through multi-dimensional detection, reducing the impact of human factors.
Smart Images

Figure CN118916444A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, system, device and storage medium for identifying buck-passing behavior. Background Art
[0002] In the public service and administrative management system, channels such as the 12345 hotline and professional service hotlines are important ways for the public to raise demands and feedback issues. The timely handling and response of public demands is an important indicator of service quality. However, in the actual operation process, some handling units have shirked their responsibilities.
[0003] At present, the method of identifying buck-passing behavior mainly relies on manual review, which requires managers to review and analyze a large amount of task order data one by one. This is not only labor-intensive and inefficient, but also highly subjective and easily affected by human factors, resulting in low accuracy of judgment. Summary of the invention
[0004] In view of this, the embodiments of the present application are committed to providing a buck-passing behavior identification method, system, device and storage medium for solving the problem of large workload and low efficiency in identifying buck-passing behavior in task order data.
[0005] According to a first aspect of the present invention, the present invention provides a method for identifying buck-passing behavior, comprising:
[0006] Obtain and analyze task order data to obtain indicators of shirking behavior, including processing time, number of processing times, negative attitudes, and irregular order returns;
[0007] Calculate the processing time abnormal value and the processing number abnormal value through the task order data, and visualize the processing time abnormal value and the processing number abnormal value through a data visualization library;
[0008] Inputting the feedback opinions of the task order data into a pre-trained natural language processing model for sentiment analysis to obtain negative sentiment in the task order data;
[0009] Analyze the flow path of the task order data and identify the task order data of irregular return orders;
[0010] According to the task order data of the irregularly returned order, it is confirmed that there is shirking behavior in the task order data.
[0011] Preferably, the calculating the processing time abnormal value and the processing number abnormal value through the task order data includes:
[0012] Calculate the processing time outlier by using an outlier detection algorithm according to the task order data;
[0013] The processing times outlier value is calculated according to the task order data by using an outlier detection algorithm.
[0014] Preferably, the visualizing the processing time abnormal value and the processing number abnormal value by using a data visualization library includes:
[0015] Draw a time series graph of the processing time and the processing times through a data visualization library;
[0016] The processing time outliers and the processing number outliers are highlighted by the time series graph.
[0017] Preferably, the natural language processing model includes:
[0018] Generate word clouds from feedback on historical task order data through WordCloud;
[0019] Performing sentiment classification on the word cloud, wherein the sentiment classification includes positive sentiment, neutral sentiment and negative sentiment;
[0020] The word cloud is used as the input of the natural language processing model, and the negative sentiment is used as the output of the natural language model.
[0021] Preferably, the analyzing the flow path of the task order data and identifying the task order data of irregular return orders includes:
[0022] The task order data flow path is identified through a rule model to obtain the task order data of the irregular return order.
[0023] Preferably, confirming that there is shirking behavior in the task order data based on the abnormal value of the processing time, the abnormal value of the number of processing times, the negative sentiment and the irregular return order includes:
[0024] When at least one of the abnormal value of the processing time, the abnormal value of the number of processing times, the negative attitude and the irregular order return exists in the task order data, it is confirmed that there is shirking behavior in the task order data.
[0025] Preferably, when at least one of the abnormal processing time, the abnormal processing times, the negative attitude and the irregular order return exists in the task order data, after confirming that the task order data contains shirking behavior, the method further includes:
[0026] The number of task orders with buck-passing behavior and the proportion of the total number of task orders in the statistical department are counted, and the department whose proportion of the number of task orders with buck-passing behavior reaches a set threshold is supervised.
[0027] According to a second aspect of the present invention, there is also provided a buck-passing behavior identification system, comprising:
[0028] An acquisition module is used to acquire and analyze task order data to obtain indicators of shirking behavior, including processing time, number of processing times, negative attitude, and irregular order returns;
[0029] A visualization module, used to calculate the processing time abnormal value and the processing number abnormal value through the task order data, and visualize the processing time abnormal value and the processing number abnormal value through a data visualization library;
[0030] The sentiment module inputs the feedback opinions of the task order data into a pre-trained natural language processing model for sentiment analysis to obtain negative sentiment in the task order data;
[0031] An identification module, which analyzes the flow path of the task order data and identifies the task order data of irregular return orders;
[0032] The confirmation module confirms that there is shirking behavior in the task order data based on the task order data of the irregular return order.
[0033] According to a third aspect of the present invention, there is also provided an electronic device, comprising a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the buck-passing behavior identification method.
[0034] According to a fourth aspect of the present invention, there is also provided a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the method for identifying buck-passing behavior.
[0035] The effects provided in the content of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:
[0036] According to the first aspect of the present invention, the present invention obtains multiple indicators of buck-passing behavior by analyzing the task list data, and calculates the value range of buck-passing behavior for each indicator, thereby judging whether buck-passing behavior exists in each task list data, and forming a multi-dimensional outlier detection method by detecting the processing time and the number of processing times, which can improve the flexibility and adaptability of outlier detection, and perform fine-grained sentiment analysis on task list feedback opinions through natural language processing technology, so that the identification of buck-passing behavior is more intelligent. Using this method to judge the buck-passing behavior existing in the task list data improves work efficiency. And judging whether buck-passing behavior exists in the task list data from four aspects can fully cover the buck-passing behavior in the task list data, ensuring the comprehensiveness and accuracy of the judgment. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 A schematic flow chart of a buck-passing behavior identification method according to an embodiment of the present invention is shown;
[0038] Figure 2 An operation flow chart of identifying buck-passing behavior according to an embodiment of the present invention is shown;
[0039] Figure 3 A schematic flow chart of a buck-passing behavior identification system according to an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0040] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0041] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.
[0042] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0043] Figure 1 FIG. 4 is a schematic flow chart of a method for identifying buck-passing behavior according to an embodiment of the present invention. Figure 1 As shown, a buck-passing behavior identification method includes:
[0044] Step S100, obtaining and analyzing task order data to obtain indicators of shirking behavior, including processing time, number of processing times, negative attitude, and irregular order returns;
[0045] Step S200, calculating the processing time abnormal value and the processing number abnormal value through the task order data, and visualizing the processing time abnormal value and the processing number abnormal value through the data visualization library;
[0046] Step S300, inputting the feedback opinions of the task order data into a pre-trained natural language processing model for sentiment analysis to obtain negative sentiment in the task order data;
[0047] Step S400, analyzing the flow path of the task order data and identifying the task order data of irregular returned orders;
[0048] Step S500, confirming that there is shirking behavior in the task order data based on the task order data of the irregularly returned order.
[0049] The present invention analyzes the task order data, obtains multiple indicators of buck-passing behavior, and calculates the value range of each indicator in the buck-passing behavior, thereby judging whether each task order data has buck-passing behavior. This method reduces the workload and improves work efficiency. Furthermore, judging whether the task order data has buck-passing behavior through four aspects can fully cover the buck-passing behavior in the task order data, and improve the accuracy of judgment.
[0050] In step S200, the abnormal value of processing time and the abnormal value of processing times are calculated by using the task list data, including:
[0051] Step S210, calculating the processing time outlier through an outlier detection algorithm according to the task order data;
[0052] Step S220, calculating the processing times outlier value through an outlier detection algorithm according to the task order data.
[0053] The outlier detection algorithm is used to calculate the normal value range of the processing time and processing times of the task order data. The outliers of the processing time and processing times of the task order data are removed within the normal value range. The time and number of times the task order data is processed are used to determine whether there is any shirking of responsibility.
[0054] First, the shirking behavior in the task order data is identified through the abnormal extension of the processing time (processing time outlier) and the frequent increase in the number of processing times (processing number outlier). The abnormal extension of the processing time requires first calculating the average processing time, the longest processing time, and the standard deviation of the processing time for each task order, and then obtaining the normal value range of the processing time based on the above three points, and then identifying the outliers of the processing time through the outlier detection algorithm, where the outlier detection algorithm includes but is not limited to statistical methods, such as Z-Score (standard score) and IQR (interquartile range), etc., or machine learning algorithms, such as isolation forest, K-Means clustering (unsupervised machine learning algorithm), etc.
[0055] The frequent increase in the number of processing times requires counting the number of times each task order is processed, identifying the task orders with frequent increases in the number of processing times, obtaining the normal value range of the number of processing times, and then identifying the abnormal values of the number of processing times through an outlier detection algorithm, where the outlier detection algorithm includes but is not limited to statistical methods such as Z-Score and IQR, or machine learning algorithms such as isolation forest and K-Means clustering, etc.
[0056] In step S200, the processing time outliers and the processing times outliers are visualized by using a data visualization library, including:
[0057] Step S230, drawing a time series graph of processing time and processing times through a data visualization library;
[0058] Step S240: highlighting abnormal values of processing duration and abnormal values of processing times through a time series graph.
[0059] Analyze the relationship between processing time, processing times and task complexity, and then use the data visualization library to draw a time series graph of processing time and processing times. Use scatter plots and box plots to display outliers of processing time and processing times, and highlight outliers to facilitate observation of abnormal performance of processing time and processing times of task order data.
[0060] In step S300, the natural language processing model includes:
[0061] Generate word clouds from feedback on historical task order data through WordCloud;
[0062] Perform sentiment classification on the word cloud, including positive sentiment, neutral sentiment and negative sentiment;
[0063] The word cloud is used as the input of the natural language processing model, and the negative sentiment is used as the output of the natural language model.
[0064] The phrases in the feedback of the task order data are used to generate a word cloud through WordCloud, and all the phrases in the word cloud are classified for sentiment. Then the natural language model is trained, and negative sentiment is used as the output of the model. The task order data is input into the model for analysis to determine whether there is a negative attitude in the task order data, and whether there is shirking behavior based on the negative attitude.
[0065] In step S400, the flow path of the task order data is analyzed to identify the task order data of the irregular returned order, including:
[0066] The task order data flow path is identified through the rule model to obtain the task order data of irregular returned orders.
[0067] At work, there is a tendency for departments to shirk responsibility. By setting rules for the rule model, we can determine whether there are irregular returns in the flow of task order data. If there are irregular returns, it can be determined that there is a tendency to shirk responsibility.
[0068] In step S500, based on the abnormal processing time, abnormal processing times, negative sentiment and irregular order returns, it is confirmed that there is shirking behavior in the task order data, including:
[0069] When at least one of the abnormal processing time, abnormal processing times, negative attitude and irregular order returns exists in the task order data, it is confirmed that there is shirking behavior in the task order data.
[0070] When at least one of the abnormal processing time, abnormal processing times, negative attitude and irregular order returns exists in the task order data, after confirming that the task order data contains shirking behavior, it also includes:
[0071] Statistics are collected on the number of task orders with buck-passing behavior and the proportion of the total number of task orders in the department, and departments are supervised if the proportion of task orders with buck-passing behavior reaches the set threshold.
[0072] In this step, when any of the abnormal processing time, abnormal processing times, negative attitude and irregular order returns appear in the task order data, the task order is marked as having shirking behavior, and a ratio is set for work orders with shirking behavior and normal work orders. When the ratio is exceeded, the department can be supervised.
[0073] Figure 2 FIG. 4 shows an operation flow chart of identifying buck-passing behavior according to an embodiment of the present invention; Figure 2As shown, in one embodiment of the present application, taking the 12345 hotline as an application scenario, the process of identifying the shirking behavior of the department processing work order data is as follows:
[0074] Obtain the task order data of the 12345 hotline, including the content of the appeal, the handling unit, the processing time and the processing results, etc. Conduct in-depth mining and analysis of the task order data to obtain potential indicators of shirking behavior in the data, including abnormal extension of processing time, frequent increase in the number of processing times, negative attitudes in feedback, and irregular returns in the process, etc.
[0075] First, the shirking behavior in the task order data is identified through the abnormal extension of the processing time and the frequent increase in the number of processing times. The abnormal extension of the processing time requires the calculation of the average processing time, the longest processing time, and the standard deviation of the processing time for each task order. Based on the above three points, the normal value range of the processing time is obtained, and then the outlier value of the processing time is identified through the outlier detection algorithm, where the outlier detection algorithm includes but is not limited to statistical methods, such as Z-Score (standard score) and IQR (interquartile range), or machine learning algorithms, such as isolation forest, K-Means clustering (unsupervised machine learning algorithm), etc.
[0076] The frequent increase in the number of processing times requires that the number of times each task order is processed is counted first, the task orders with frequently increased processing times are identified, the normal value range of the processing times is obtained, and then the outliers of the processing times are identified through an outlier detection algorithm, where the outlier detection algorithm includes but is not limited to statistical methods, such as Z-Score and IQR, or machine learning algorithms, such as isolation forest, K-Means clustering, etc.
[0077] Visualization is performed through a data visualization library based on the processing time and the number of processing times. The data visualization library includes but is not limited to Matplotlib (drawing library) or Seaborn (data visualization library), etc., to form a time series graph. Trends and abnormal points are observed through the time series graph, and the relationship between the processing time, the number of processing times and the task complexity is displayed through scatter plots and box plots, highlighting abnormal values.
[0078] Then, we identify the negative attitudes in the feedback and classify the feedback into positive, neutral, and negative. We first generate a word cloud of the feedback using WordCloud to visually display high-frequency words, especially negative words, and then train a sentiment classification model to identify negative feedback. The sentiment classification model uses natural language processing technology, such as the Natural Language Toolkit (NLTK), spaCy (a fast, industrial-grade natural language processing library), and TextBlob (a simplified natural language processing library).
[0079] Finally, identify irregular order returns in process operations, define the standards for irregular order returns through business rules, and identify irregular order return behaviors in the task order flow path through rule models, where the rule model includes but is not limited to rule engines or custom functions.
[0080] Based on the abnormal extension of processing time, frequent increase in processing times, negative attitude in feedback, irregular order returns and other shirking behaviors in the task order data, multi-dimensional behavior labels are marked for the task orders.
[0081] Based on the multi-dimensional behavior labels on the task orders, we conduct an in-depth analysis of the handling departments, count the percentage of work orders and task orders with the shirking behavior labels by department, and supervise and urge the departments to handle the work.
[0082] Figure 3 FIG. 4 shows a schematic flow chart of a buck-passing behavior identification system according to an embodiment of the present invention. Figure 3 As shown, the present application also provides a buck-passing behavior identification system, including:
[0083] The acquisition module 510 is used to acquire and analyze the task order data to obtain indicators of the shirking behavior, including processing time, processing times, negative attitude and irregular order returns;
[0084] A visualization module 520 is used to calculate the processing time abnormal value and the processing number abnormal value through the task order data, and visualize the processing time abnormal value and the processing number abnormal value through the data visualization library;
[0085] The sentiment module 530 inputs the feedback opinions of the task order data into a pre-trained natural language processing model for sentiment analysis to obtain negative sentiment in the task order data;
[0086] Identification module 540, analyzing the flow path of task order data and identifying task order data of irregular returned orders;
[0087] The confirmation module 550 confirms whether there is any shirking of responsibility in the task order data according to the task order data of the irregular returned order.
[0088] In particular, the present application also provides an electronic device, the electronic device includes a processor and a memory, the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the buck-passing behavior identification method, for example, executing the above-described Figure 1 The method comprises steps S100 to S500.
[0089] In particular, the present application also provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the buck-passing behavior identification method, for example, to execute the above-described Figure 1 The method comprises steps S100 to S500.
[0090] The processor is used to complete various control logics of the device, which can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISCMachine) or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination of these components. In addition, the processor can also be any traditional processor, microprocessor or state machine. The processor can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP and / or any other such configuration.
[0091] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0092] As an example, non-volatile storage media can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) as external cache memory. By way of illustration and not limitation, RAM can be obtained in many forms such as synchronous RAM (SRAM), dynamic RAM, (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The disclosed memory components or memories of the operating environment described herein are intended to include one or more of these and / or any other suitable types of memory.
[0093] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium, and the computer program can include the processes of the above-mentioned method embodiments when executed. The storage medium can be a memory, a disk, a floppy disk, a flash memory, an optical storage device, etc.
[0094] The above are only some specific implementation methods of the present application, and any other improvements made based on the concept of the present application are deemed to be within the protection scope of the present application.
Claims
1. A method for identifying buck-passing behavior, characterized in that: include: Obtain and analyze task order data to obtain indicators of shirking behavior, including processing time, number of processing times, negative attitudes, and irregular order returns; Calculate the processing time abnormal value and the processing number abnormal value through the task order data, and visualize the processing time abnormal value and the processing number abnormal value through a data visualization library; Inputting the feedback opinions of the task order data into a pre-trained natural language processing model for sentiment analysis to obtain negative sentiment in the task order data; Analyze the flow path of the task order data and identify the task order data of irregular return orders; Based on the abnormal processing time, the abnormal processing times, the negative emotions and the irregular order returns, it is confirmed that there is shirking behavior in the task order data.
2. The method according to claim 1, characterized in that The calculating the processing time abnormal value and the processing number abnormal value by using the task list data includes: Calculate the processing time outlier by using an outlier detection algorithm according to the task order data; The processing times outlier value is calculated according to the task order data by using an outlier detection algorithm.
3. The method according to claim 2, characterized in that The visualizing the processing time abnormal value and the processing number abnormal value by using a data visualization library includes: Draw a time series graph of the processing time and the processing times through a data visualization library; The processing time outliers and the processing number outliers are highlighted by the time series graph.
4. The method according to claim 3, characterized in that: The pre-training of the natural language processing model includes: Generate word clouds from feedback on historical task order data through WordCloud; Performing sentiment classification on the word cloud, wherein the sentiment classification includes positive sentiment, neutral sentiment and negative sentiment; The word cloud is used as the input of the natural language processing model, and the negative sentiment is used as the output of the natural language model.
5. The method according to claim 1, characterized in that The analyzing the flow path of the task order data and identifying the task order data of irregular return orders includes: The task order data flow path is identified through a rule model to obtain the task order data of the irregular return order.
6. The method according to claim 1, characterized in that The step of confirming that there is shirking behavior in the task order data based on the abnormal value of the processing time, the abnormal value of the number of processing times, the negative sentiment, and the irregular order return includes: When at least one of the abnormal value of the processing time, the abnormal value of the number of processing times, the negative attitude and the irregular order return exists in the task order data, it is confirmed that there is shirking behavior in the task order data.
7. The method according to claim 6, characterized in that When at least one of the abnormal processing time, the abnormal processing times, the negative attitude and the irregular order return exists in the task order data, after confirming that the task order data contains shirking behavior, the method further includes: The number of task orders with buck-passing behavior and the proportion of the total number of task orders in the statistical department are counted, and the department whose proportion of the number of task orders with buck-passing behavior reaches a set threshold is supervised.
8. A buck-passing behavior recognition system, characterized in that: include: An acquisition module is used to acquire and analyze task order data to obtain indicators of shirking behavior, including processing time, number of processing times, negative attitude, and irregular order returns; A visualization module, used to calculate the processing time abnormal value and the processing number abnormal value through the task order data, and visualize the processing time abnormal value and the processing number abnormal value through a data visualization library; An emotion module, used for inputting the feedback opinions of the task order data into a pre-trained natural language processing model for emotion analysis, and obtaining negative emotions in the task order data; An identification module, used to analyze the flow path of the task order data and identify the task order data of irregular return orders; The confirmation module is used to confirm whether there is shirking behavior in the task order data according to the task order data of the irregular return order.
9. An electronic device, characterized in that: The electronic device includes a processor and a memory, the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the buck-passing behavior identification method described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by the processor, the computer-executable instructions prompt the processor to implement the buck-passing behavior identification method described in any one of claims 1 to 7.
Citation Information
Patent Citations
Service state processing method and device, model training method and device, equipment and storage medium
CN111897705A
Public transportation passenger satisfaction evaluation method and system based on public opinion knowledge graph
CN111914096A
Risk control rule tuning method and system
CN112686521A
Business index abnormity detection method and device, electronic equipment and storage medium
CN113822366A
Natural language computer processing system
CN118468867A
