CN111062519A - Method and device for sensing customer satisfaction - Google Patents
Method and device for sensing customer satisfaction Download PDFInfo
- Publication number
- CN111062519A CN111062519A CN201911182946.3A CN201911182946A CN111062519A CN 111062519 A CN111062519 A CN 111062519A CN 201911182946 A CN201911182946 A CN 201911182946A CN 111062519 A CN111062519 A CN 111062519A
- Authority
- CN
- China
- Prior art keywords
- sample attribute
- use data
- sample
- probability
- change rate
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention belongs to the technical field of information processing, and particularly discloses a method and a device for sensing customer satisfaction. The method comprises the following steps: acquiring customer use data; preprocessing customer use data to obtain target customer use data; obtaining a sample attribute and a corresponding sample attribute value based on the target customer usage data; obtaining a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and the corresponding sample attribute value; classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first set of samples and a second set of samples; respectively calculating and obtaining a first probability and a second probability by using a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples and the second group of samples; and determining the customer satisfaction according to the first probability and the second probability. The method can accurately predict the satisfaction degree of the user so as to be used for the client service department to develop effective client maintenance work and reduce client loss.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to a method and a device for sensing customer satisfaction.
Background
The customer satisfaction can provide accurate customer feedback information for the merchant, improve service or adjust marketing strategies in time, provide high-quality service for the customer and obtain rich profits.
Currently, the method of sensing customer satisfaction mainly collects customer consultation, advice or complaint through a "hotline" phone and then derives customer satisfaction based on the collected information. However, the customer hotline is a post-evaluation mechanism, and the receipt of a telephone complaint often indicates that a very poor user experience has been caused to the user and the image of the enterprise is adversely affected. In addition, there is a lot of invalid information in the complaints and suggestions of the customers due to the age, education level, expressive power, etc. of the customers. Thus, the customer hotline has the problems of poor timeliness and large amount of invalid information, resulting in the inability of the merchant to improve service or adjust marketing strategies in a timely manner.
Disclosure of Invention
Therefore, the invention provides a method and a device for sensing customer satisfaction, which aim to solve the problem that marketing strategies cannot be adjusted and services cannot be improved in time due to poor timeliness and a large amount of invalid information of a customer hot line in the prior art.
In order to achieve the above object, a first aspect of the present invention provides a method of sensing customer satisfaction, comprising:
acquiring customer use data, wherein the customer use data is the use data of a communication service process used by a user;
preprocessing the customer use data to obtain target customer use data;
obtaining a sample attribute and a corresponding sample attribute value based on the target customer usage data;
obtaining a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and the corresponding sample attribute value;
classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first set of samples and a second set of samples;
calculating to obtain a first probability by using a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples, and calculating to obtain a second probability by using the Bayesian formula based on the change rate of the attribute value of each sample in the second group of samples;
and determining the customer satisfaction according to the first probability and the second probability.
Wherein the preprocessing the customer usage data to obtain target customer usage data comprises:
extracting client use data with the call duration less than the preset duration from the client use data to obtain first preprocessing use data, wherein the first preprocessing use data comprises a called number and the call duration;
counting the use data of which the number of times of calls with the called number exceeds a preset number of times of calls within a preset time period from the first pre-processed use data, and obtaining a first statistical value;
counting the use data with the average flow velocity smaller than the preset flow velocity in a preset time period from the first preprocessing use data, and obtaining a second statistical value;
determining an acquisition target based on the first statistical value and the second statistical value;
and acquiring the use data of the acquired target to obtain the use data of the target customer.
Wherein the determining an information collection target based on the first statistical value and the second statistical value comprises:
calculating the sum of the first statistical value and the second statistical value to obtain a statistical total value;
and judging whether the total statistical value is greater than a preset threshold value, and if the total statistical value is greater than the preset threshold value, taking the customer use data as a collection target.
Wherein the obtaining of the sample attribute value change rate corresponding to the sample attribute based on the sample attribute value corresponding to the sample attribute includes:
obtaining a first sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client usage data in a first preset period;
obtaining a second sample attribute value based on a sample attribute value corresponding to each sample attribute in the customer usage data in a second preset period, wherein the first preset period is shorter than the second preset period;
obtaining a sample attribute value change rate based on the first sample attribute value and the second sample attribute value, wherein the sample attribute value change rate comprises a call frequency change rate, a call duration change rate, a flow rate change rate, a complaint frequency change rate and a non-service frequency change rate.
Wherein, the obtaining of the first probability by respectively calculating the change rate of the attribute value of each sample in the first group of samples by using a Bayesian formula comprises:
obtaining a first probability by using a Bayesian formula based on the satisfied probability corresponding to each sample attribute in the first group of samples, the joint probability of each sample attribute and the total satisfied probability;
the obtaining of the second probability by respectively calculating the change rate of the attribute value of each sample in the second group of samples by using a Bayesian formula includes:
and obtaining a second probability by using a Bayesian formula based on the satisfaction probability corresponding to each sample attribute, the joint probability of each sample attribute and the total satisfaction probability in the second group of samples.
Wherein the determining satisfaction according to the first probability and the second probability comprises:
comparing the first probability and the second probability;
and selecting the person with the large probability value as the customer satisfaction.
In order to achieve the above object, a second aspect of the present invention provides an apparatus for sensing customer satisfaction, comprising:
the data acquisition unit is used for acquiring client use data, and the client use data is use data of a communication service process used by a user;
the preprocessing unit is used for preprocessing the client use data to obtain target client use data;
an extracting unit for extracting a sample attribute and a corresponding sample attribute value based on the target customer usage data;
a sample attribute value change rate obtaining unit, configured to obtain a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and a corresponding sample attribute value;
a classification unit for classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first set of samples and a second set of samples;
the probability calculation unit is used for calculating and obtaining a first probability by utilizing a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples and calculating and obtaining a second probability by utilizing the Bayesian formula based on the change rate of the attribute value of each sample in the second group of samples;
and the satisfaction confirming unit is used for determining the customer satisfaction according to the first probability and the second probability.
Wherein the preprocessing unit comprises:
the selection module is used for extracting client use data with the call duration less than the preset duration from the client use data to obtain first preprocessing use data, and the first preprocessing use data comprises a called number and the call duration;
the first statistical module is used for counting the use data of which the number of times of conversation with the called number exceeds the preset number within a preset time period from the first preprocessing use data and obtaining a first statistical value;
the second counting module is used for counting the use data of which the average flow speed is less than the preset flow speed within a preset time period from the first preprocessing use data and obtaining a second counting value;
the acquisition target confirming module is used for confirming an acquisition target based on the first statistical value and the second statistical value;
and the acquisition module acquires the use data of the acquired target to obtain the use data of the target customer.
Wherein, the collection target confirmation module comprises:
the calculation submodule is used for calculating the sum of the first statistical value and the second statistical value to obtain a statistical total value;
and the judging submodule is used for judging whether the statistical total value is greater than a preset threshold value or not, and if the statistical total value is greater than the preset threshold value, the customer use data is an acquisition target.
Wherein, the sample attribute value change rate obtaining unit includes:
the first extraction module is used for obtaining a first sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client use data in a first preset period;
the second extraction module is used for obtaining a second sample attribute value based on a sample attribute value corresponding to each sample attribute in the target customer usage data within a second preset period, wherein the first preset period is smaller than the second preset period;
and the sample attribute value change rate acquisition module is used for acquiring a sample attribute value change rate based on the first sample attribute value and the second sample attribute value, wherein the sample attribute value change rate comprises a call frequency change rate, a call duration change rate, a flow rate change rate, a complaint frequency change rate and a non-service frequency change rate.
The invention has the following advantages:
the method for sensing the customer satisfaction degree provided by the embodiment of the invention comprises the steps of preprocessing customer use data to obtain target customer use data; obtaining sample attributes and corresponding sample attribute values from the target customer usage data; then obtaining the change rate of the sample attribute value according to the sample attribute and the corresponding sample attribute value; classifying the target client use data based on satisfaction and dissatisfaction to obtain a first group of samples and a second group of samples, respectively calculating to obtain a first probability based on the attribute value change rate of each sample in the first group of samples by using a Bayesian formula, and respectively calculating to obtain a second probability based on the attribute value change rate of each sample in the second group of samples by using the Bayesian formula; and finally, determining the customer satisfaction according to the first probability and the second probability, thereby accurately predicting the user satisfaction to obtain an accurate target customer group so that a customer service department can develop effective customer maintenance work and reduce customer loss.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
FIG. 1 is a flow chart of a method for sensing customer satisfaction provided by an embodiment of the present invention;
FIG. 2 is a flow chart of a preprocessing step in a method for sensing customer satisfaction according to an embodiment of the present invention;
FIG. 3 is a block diagram of an apparatus for sensing customer satisfaction according to an embodiment of the present invention;
fig. 4 is a block diagram of a preprocessing unit according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
The method and the device for perceiving the customer satisfaction degree can be applied to communication operators and can also be applied to other industries such as service and sale and the like to predict the customer satisfaction degree, so that the operators can adjust marketing strategies in time and improve services.
Fig. 1 is a flowchart of a method for sensing customer satisfaction according to this embodiment. The present embodiment is mainly described for a telecom operator as an example. As shown in fig. 1, a method of sensing customer satisfaction includes:
step S101, customer usage data is acquired.
Wherein, the client use data is the use data of the communication service process used by the user. The customer usage data includes, but is not limited to, customer voice, data services, etc.
The customer voice includes, but is not limited to, a customer number (calling number), a party number (called number), a call duration, a start time, and the like. The duration of the call can be accurate to seconds, and the start time is also calculated in seconds. Table 1 illustrates an example of a portion of customer voice data.
TABLE 1
| Customer number | Number of the other party | Duration of call | Starting time |
| 18600000000 | 18600000001 | 9:54:34 | 16:08:01 |
| 18600000000 | 18600000001 | 9:38:20 | 16:27:17 |
| 18600000000 | 18600000001 | 10:06:08 | 10:01:35 |
The data traffic includes, but is not limited to, the number of the client, the duration, the type of the network, the total flow rate, etc. The network types include, but are not limited to, 2G, 3G, 4G, and future 5G mobile communication network types. Table 2 exemplarily shows a part of the data traffic data.
TABLE 2
| Customer number | Duration of time | Network type | Total flow Rate/KB |
| 18600000000 | 15:34 | 4G | 61440.29 |
| 18600000000 | 2:20 | 4G | 2847.94 |
| 18600000000 | 1:08 | 4G | 2471.9 |
In one embodiment, historical customer usage data for the last month or more is obtained from the operator's Base Station Subsystem (BSS) or centralized service support System (CBSS).
And step S102, preprocessing the customer use data to obtain target customer use data.
The client usage data obtained in step S101 is large, and affects the subsequent processing speed. Therefore, the customer use data which is helpful for predicting the customer satisfaction needs to be selectively extracted from the customer use data, the range of the customer use data is narrowed, and the data which has little or no effect on the customer satisfaction prediction is excluded.
In one embodiment, as shown in FIG. 2, the pre-processing consists essentially of:
step S201, extracting the customer use data with the call duration less than the preset duration from the customer use data, and obtaining first preprocessing use data.
The first preprocessing usage data comprises a called number and a call duration.
In one embodiment, the call duration is utilized to compress the customer usage data. Specifically, the client use data with the call duration being greater than or equal to the preset duration is excluded, and only the client use data with the call duration being less than the preset duration is reserved. First pre-processing usage data is obtained through selection of the customer usage data.
It should be noted that the preset duration may be set by the operator according to the needs, or may be set by a third party of the satisfaction survey. For example, the preset time period is set to 10 seconds. And selecting the client use data with the call duration less than 10 seconds to obtain first preprocessing use data.
Step S202, counting the use data of which the number of times of calling with the called number exceeds the preset number of times of calling in a preset time period from the first preprocessing use data, and obtaining a first statistical value.
The preset time period and the preset number of times of call can be set by an operator or can be set by a third party for satisfaction survey. The preset time period is set to one or two weeks, and may be one month or longer. The preset number of calls may be set to 3, 5, or other numbers.
In one embodiment, for each piece of customer usage data, according to the fields of the call duration and the number of the opposite party, the number and duration of calls with the same number of the opposite party in continuous time are found, and if the number of calls is more than 3, the piece of customer usage data is recorded. Then, counting the use data of which the number of calls with the called number exceeds 3 times and the call duration is less than 10 seconds within one week to obtain a first statistical value which is recorded as sumN.
Step S203, counting, from the first pre-processed usage data, usage data with a mean flow rate less than a preset flow rate within a preset time period, and obtaining a second statistical value.
The preset time period and the preset number of times of call can be set by an operator or a third party for satisfaction survey. The preset time period is set to one or two weeks, and may be one month or longer. The preset flow rate is set to 100k/s, 200k/s or 500k/s, but may be set to other values.
In one embodiment, usage data having a mean flow rate of less than 100k/s over a week is counted from the first pre-processed usage data and a second statistical value, denoted sumV, is obtained.
And step S204, determining an acquisition target based on the first statistical value and the second statistical value.
In one embodiment, determining the acquisition target based on the first statistical value and the second statistical value comprises: calculating the sum of the first statistical value and the second statistical value to obtain a statistical total value; and judging whether the statistical total value is greater than a preset threshold value, and if the statistical total value is greater than the preset threshold value, taking the customer use data as a collection target.
The preset threshold value can be set by an operator or a third party for satisfaction survey.
In one embodiment, if the statistical total is greater than a preset threshold, the customer uses the data as a collection target. And determining an acquisition target according to the client number in the client use data, and then tracking the signaling.
And step S205, acquiring the use data of the target to obtain the use data of the target customer.
The collection amount of the data used by the customer is reduced through the steps S201 to S205, which is beneficial to improving the subsequent processing efficiency.
Step S103, obtaining sample attributes and corresponding sample attribute values based on the target customer usage data.
Sample attributes include, but are not limited to, number of complaints, number of out-of-service times, number of calls, length of call, and flow rate, among others. The sample attribute value is a numerical value corresponding to the sample attribute, for example, the sample attribute value of the number of complaints is the specific number of complaints, the sample attribute value of the number of non-service is the specific number of non-service, the sample attribute value of the number of calls is the specific number of calls, the sample attribute value of the call duration is the specific duration of calls, and the sample attribute value of the flow rate is the specific speed of the network data.
And step S104, obtaining the sample attribute value change rate corresponding to the sample attribute based on the sample attribute and the corresponding sample attribute value.
Wherein, the sample attribute value change rate is a prediction factor of the prediction satisfaction, including but not limited to call number change rate, call duration change rate, flow rate change rate, complaint number change rate and out-of-service number change rate.
In one embodiment, the sample property value change rate is obtained by: obtaining a first sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client use data in a first preset period; obtaining a second sample attribute value based on the sample attribute value corresponding to each sample attribute in the customer use data in a second preset period, wherein the first preset period is smaller than the second preset period; a sample attribute value change rate is obtained based on the first sample attribute value and the second sample attribute value.
In one embodiment, the first predetermined period is one week and the second predetermined period is one month. Customer usage data for the target is collected over seven days (first period T1) and a first sample attribute value are obtained. Meanwhile, the client usage data of the collection target for 70 days (second period T2) is collected, and the second sample attribute value are obtained. A sample attribute value change rate is obtained based on the first sample attribute value and the second sample attribute value.
For example, the average number of calls within the first period T1 is calculated: the total number of calls is T1/T2, wherein the total number of calls is the number of calls in the second period T2.
The average total call duration within the first period T1 is calculated: the total duration of the call is multiplied by T1/T2, wherein the total duration of the call is the total duration of the call in the second period T2.
The average flow rate over the first period T1 is calculated: the total flow is T1/T2, where the total flow is the total flow of data during the second period T2.
Since the second period T2 includes a plurality of first periods T1, the average call frequency, the average total call duration, and the average flow rate of the other periods T1 in the second period T2 are calculated in the same manner, and then the sample attribute value change rate is calculated, that is, the call frequency change rate, the call duration change rate, the flow rate change rate, the complaint frequency change rate, and the out-of-service frequency change rate are obtained.
Step S105, classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first group of samples and a second group of samples.
And obtaining a first probability by using a Bayesian formula based on the satisfactory probability corresponding to each sample attribute in the first group of samples, the joint probability of each sample attribute and the total satisfactory probability. And obtaining a second probability by using a Bayesian formula based on the satisfaction probability corresponding to each sample attribute, the joint probability of each sample attribute and the total satisfaction probability in the second group of samples by using the same method.
It should be noted that the first group of samples and the second group of samples belong to samples in the same period T1, and the number of samples is the same, and the satisfaction of the first group of samples is "unsatisfactory" and the satisfaction of the second group of samples is "satisfactory".
Tables 3 and 4 schematically show a first set of samples and a second set of samples, wherein each sample comprises a sample property change rate.
TABLE 3 sample Property Change rates for the first set of samples
TABLE 4 sample Property Change rates for the second set of samples
Step S106, a first probability is obtained through calculation by using a Bayes formula based on the change rate of the attribute value of each sample in the first group of samples, and a second probability is obtained through calculation by using the Bayes formula based on the change rate of the attribute value of each sample in the second group of samples.
And in the first group of samples, obtaining a probability distribution table of the sample attribute change rate according to the probability distribution of the complaint times, the non-service times, the call time change rate, the call duration change rate and the flow change rate. For example, the number of complaints is counted and the probability of the number of complaints is calculated as shown in Table 5.
TABLE 5 probability distribution Table for complaint frequency of first set of samples
In the second group of samples, a probability distribution table of the sample attribute change rate is obtained according to the probability distribution of the complaint times, the non-service times, the call time change rate, the call duration change rate, and the traffic change rate, as shown in table 6.
TABLE 6 probability distribution Table for complaint times for the second set of samples
The first probability is calculated by the formula:
P(A|X)＝P(A|x1,x2,x3,x4,x5)
＝P(x1,x2,x3,x4,x5|A)P(A)/P(x1,x2,x3,x4,x5)
＝P(x1|A)P(x2|A)P(x3|A)P(x4|A)P(x5|A)P(A)/P(x1,x2,x3,x4,x5)
＝P(x1|A)P(x2|A)P(x3|A)P(x4|A)P(x5|A)P(A)/P(X)
where a is dissatisfied, P (x1| a) indicates the probability that the dissatisfied sample property change rate is x1, P (a) indicates the dissatisfaction rates for all sample property change rates, and P (x) indicates the joint probability.
The second probability is calculated as:
P(B|X)
＝P(B|x1,x2,x3,x4,x5)
＝P(x1,x2,x3,x4,x5|B)P(B)/P(x1,x2,x3,x4,x5)
＝P(x1|B)P(x2|B)P(x3|B)P(x4|B)P(x5|B)P(B)/P(x1,x2,x3,x4,x5)
＝P(x1|B)P(x2|B)P(x3|B)P(x4|B)P(x5|B)P(B)/P(X)
where B is satisfied, P (x1| B) represents the probability that the satisfied sample property change rate is x1, P (B) represents the satisfaction rate for all sample property change rates, and P (x) represents the joint probability.
In the calculation of the first probability and the second probability, since the number of the first group of samples and the second group of samples is the same, p (a) and p (b) are included.
P (X) represents the joint probability of the rate of change of the attribute of each sample, i.e.
P(X)＝P(x1)P(x2)P(x3)P(x4)P(x5)
For any customer, the values of p (x) in the equations, which can be calculated from the values of the sample property change rates x1, x2, x3, x4, x5, are equal.
Therefore, p (a)/p (x) ═ p (b)/p (x).
And step S107, determining the customer satisfaction according to the first probability and the second probability.
Comparing the first probability P (A | X) and the second probability P (B | X), wherein the larger one is the prediction result, namely, if the first probability P (A | X) is larger than the second probability P (B | X), the customer satisfaction is represented as dissatisfaction; if the first probability P (A | X) is less than the second probability P (B | X), then the customer satisfaction is indicated as satisfactory; if the first probability P (a | X) and the second probability P (B | X) are equal, the operator may directly determine satisfaction or dissatisfaction, or may determine the probability based on other factors.
The method for sensing the customer satisfaction degree provided by the embodiment of the invention comprises the steps of preprocessing customer use data to obtain target customer use data; obtaining sample attributes and corresponding sample attribute values from the target customer usage data; then obtaining the change rate of the sample attribute value according to the sample attribute and the corresponding sample attribute value; classifying target customer use data based on satisfaction and dissatisfaction to obtain a first group of samples and a second group of samples, respectively calculating to obtain a first probability based on the attribute value change rate of each sample in the first group of samples by using a Bayesian formula, and respectively calculating to obtain a second probability based on the attribute value change rate of each sample in the second group of samples by using the Bayesian formula; and finally, determining the customer satisfaction according to the first probability and the second probability so as to predict the satisfaction of the mastered user and obtain an accurate target customer group, so that the customer service department can develop effective customer maintenance work and reduce customer loss.
The steps of the above methods are divided for clarity, and the implementation may be combined into one step or split some steps, and the steps are divided into multiple steps, so long as the same logical relationship is included, which are all within the protection scope of the present patent; it is within the scope of the patent to add insignificant modifications to the algorithms or processes or to introduce insignificant design changes to the core design without changing the algorithms or processes.
Fig. 3 is a block diagram of an apparatus for sensing customer satisfaction provided by an embodiment of the present invention. The device for sensing the customer satisfaction corresponds to the method for sensing the customer satisfaction, and comprises the following steps:
a data acquiring unit 301, configured to acquire client usage data, which is usage data of a communication service process used by a user.
And a preprocessing unit 302, configured to preprocess the client usage data to obtain target client usage data.
An extracting unit 303, configured to extract a sample attribute and a corresponding sample attribute value based on the target customer usage data.
A sample attribute value change rate obtaining unit 304, configured to obtain a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and the corresponding sample attribute value.
A classification unit 305 for classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first set of samples and a second set of samples.
And the probability calculating unit 306 is configured to calculate and obtain a first probability by using a bayesian formula based on the change rate of the attribute value of each sample in the first group of samples, and calculate and obtain a second probability by using the bayesian formula based on the change rate of the attribute value of each sample in the second group of samples.
A satisfaction confirming unit 307, configured to determine the customer satisfaction according to the first probability and the second probability.
In one embodiment, as shown in fig. 4, the preprocessing unit includes:
the selecting module 401 is configured to extract, from the client usage data, client usage data with a call duration less than a preset duration, and obtain first pre-processing usage data, where the first pre-processing usage data includes a called number and the call duration.
A first statistical module 402, configured to count, from the first pre-processed usage data, usage data in which the number of calls with the called number exceeds a preset number within a preset time period, and obtain a first statistical value.
A second counting module 403, configured to count, from the first pre-processing usage data, usage data with a mean flow rate smaller than a preset flow rate within a preset time period, and obtain a second statistical value.
The acquisition target determination module 404 determines an acquisition target based on the first statistical value and the second statistical value. Acquisition target confirmation module comprising: the calculation submodule is used for calculating the sum of the first statistical value and the second statistical value to obtain a statistical total value; and the judging submodule is used for judging whether the statistical total value is greater than a preset threshold value or not, and if the statistical total value is greater than the preset threshold value, the customer use data is taken as an acquisition target.
And the acquisition module 405 acquires the use data of the target to obtain the use data of the target customer.
In one embodiment, the sample attribute value change rate acquisition unit includes: the first extraction module is used for obtaining a first sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client use data in a first preset period; the second extraction module is used for obtaining a second sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client use data in a second preset period, wherein the first preset period is smaller than the second preset period; and the sample attribute value change rate acquisition module is used for acquiring a sample attribute value change rate based on the first sample attribute value and the second sample attribute value, wherein the sample attribute value change rate comprises a call frequency change rate, a call duration change rate, a flow rate change rate, a complaint frequency change rate and a non-service frequency change rate.
In the device for sensing customer satisfaction provided by the embodiment of the invention, an acquisition unit is used for acquiring customer use data, a preprocessing unit is used for preprocessing the customer use data, an extraction unit is used for extracting a sample attribute and a corresponding sample attribute value, and a sample attribute value change rate acquisition unit is used for acquiring a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and the corresponding sample attribute value; the probability calculation unit is used for respectively calculating to obtain a first probability by utilizing a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples and respectively calculating to obtain a second probability by utilizing the Bayesian formula based on the change rate of the attribute value of each sample in the second group of samples; and finally, determining the customer satisfaction by the satisfaction confirming unit according to the first probability and the second probability so as to predict and master the satisfaction of the user and obtain an accurate target customer group, so that the customer service department can develop effective customer maintenance work and reduce customer loss.
Each module in the present embodiment is a logical module, and in practical applications, one logical unit may be one physical unit, may be a part of one physical unit, or may be implemented by a combination of a plurality of physical units. In addition, in order to highlight the innovative part of the present invention, elements that are not so closely related to solving the technical problems proposed by the present invention are not introduced in the present embodiment, but this does not indicate that other elements are not present in the present embodiment.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (10)
1. A method of sensing customer satisfaction, comprising:
acquiring customer use data, wherein the customer use data is the use data of a communication service process used by a user;
preprocessing the customer use data to obtain target customer use data;
obtaining a sample attribute and a corresponding sample attribute value based on the target customer usage data;
obtaining a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and the corresponding sample attribute value;
classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first set of samples and a second set of samples;
calculating to obtain a first probability by using a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples, and calculating to obtain a second probability by using the Bayesian formula based on the change rate of the attribute value of each sample in the second group of samples;
and determining the customer satisfaction according to the first probability and the second probability.
2. The method of claim 1, wherein said preprocessing said customer usage data to obtain target customer usage data comprises:
extracting client use data with the call duration less than the preset duration from the client use data to obtain first preprocessing use data, wherein the first preprocessing use data comprises a called number and the call duration;
counting the use data of which the number of times of calls with the called number exceeds a preset number of times of calls within a preset time period from the first pre-processed use data, and obtaining a first statistical value;
counting the use data with the average flow velocity smaller than the preset flow velocity in a preset time period from the first preprocessing use data, and obtaining a second statistical value;
determining an acquisition target based on the first statistical value and the second statistical value;
and acquiring the use data of the acquired target to obtain the use data of the target customer.
3. The method of claim 2, wherein determining an information collection goal based on the first statistical value and the second statistical value comprises:
calculating the sum of the first statistical value and the second statistical value to obtain a statistical total value;
and judging whether the total statistical value is greater than a preset threshold value, and if the total statistical value is greater than the preset threshold value, taking the customer use data as a collection target.
4. The method of claim 2, wherein obtaining the sample attribute value variation rate corresponding to the sample attribute based on the sample attribute value corresponding to the sample attribute comprises:
obtaining a first sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client usage data in a first preset period;
obtaining a second sample attribute value based on a sample attribute value corresponding to each sample attribute in the customer usage data in a second preset period, wherein the first preset period is shorter than the second preset period;
obtaining a sample attribute value change rate based on the first sample attribute value and the second sample attribute value, wherein the sample attribute value change rate comprises a call frequency change rate, a call duration change rate, a flow rate change rate, a complaint frequency change rate and a non-service frequency change rate.
5. The method of claim 4, wherein the obtaining the first probability by separately calculating using a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples comprises:
obtaining a first probability by using a Bayesian formula based on the satisfied probability corresponding to each sample attribute in the first group of samples, the joint probability of each sample attribute and the total satisfied probability;
the obtaining of the second probability by respectively calculating the change rate of the attribute value of each sample in the second group of samples by using a Bayesian formula includes:
and obtaining a second probability by using a Bayesian formula based on the satisfaction probability corresponding to each sample attribute, the joint probability of each sample attribute and the total satisfaction probability in the second group of samples.
6. The method of claim 1, wherein said determining a satisfaction from said first and second probabilities comprises:
comparing the first probability and the second probability;
and selecting the person with the large probability value as the customer satisfaction.
7. An apparatus for sensing customer satisfaction, comprising:
the data acquisition unit is used for acquiring client use data, and the client use data is use data of a communication service process used by a user;
the preprocessing unit is used for preprocessing the client use data to obtain target client use data;
an extracting unit for extracting a sample attribute and a corresponding sample attribute value based on the target customer usage data;
a sample attribute value change rate obtaining unit, configured to obtain a sample attribute value change rate corresponding to the sample attribute based on the sample attribute and a corresponding sample attribute value;
a classification unit for classifying the target customer usage data based on satisfaction and dissatisfaction in the sample attributes to obtain a first set of samples and a second set of samples;
the probability calculation unit is used for calculating and obtaining a first probability by utilizing a Bayesian formula based on the change rate of the attribute value of each sample in the first group of samples and calculating and obtaining a second probability by utilizing the Bayesian formula based on the change rate of the attribute value of each sample in the second group of samples;
and the satisfaction confirming unit is used for determining the customer satisfaction according to the first probability and the second probability.
8. The apparatus of claim 7, wherein the pre-processing unit comprises:
the selection module is used for extracting client use data with the call duration less than the preset duration from the client use data to obtain first preprocessing use data, and the first preprocessing use data comprises a called number and the call duration;
the first statistical module is used for counting the use data of which the number of times of conversation with the called number exceeds the preset number within a preset time period from the first preprocessing use data and obtaining a first statistical value;
the second counting module is used for counting the use data of which the average flow speed is less than the preset flow speed within a preset time period from the first preprocessing use data and obtaining a second counting value;
the acquisition target confirming module is used for confirming an acquisition target based on the first statistical value and the second statistical value;
and the acquisition module acquires the use data of the acquired target to obtain the use data of the target customer.
9. The apparatus of claim 8, wherein the acquisition target confirmation module comprises:
the calculation submodule is used for calculating the sum of the first statistical value and the second statistical value to obtain a statistical total value;
and the judging submodule is used for judging whether the statistical total value is greater than a preset threshold value or not, and if the statistical total value is greater than the preset threshold value, the customer use data is an acquisition target.
10. The apparatus according to claim 7, wherein the sample attribute value change rate obtaining unit includes:
the first extraction module is used for obtaining a first sample attribute value based on the sample attribute value corresponding to each sample attribute in the target client use data in a first preset period;
the second extraction module is used for obtaining a second sample attribute value based on a sample attribute value corresponding to each sample attribute in the target customer usage data within a second preset period, wherein the first preset period is smaller than the second preset period;
and the sample attribute value change rate acquisition module is used for acquiring a sample attribute value change rate based on the first sample attribute value and the second sample attribute value, wherein the sample attribute value change rate comprises a call frequency change rate, a call duration change rate, a flow rate change rate, a complaint frequency change rate and a non-service frequency change rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911182946.3A CN111062519A (en) | 2019-11-27 | 2019-11-27 | Method and device for sensing customer satisfaction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911182946.3A CN111062519A (en) | 2019-11-27 | 2019-11-27 | Method and device for sensing customer satisfaction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111062519A true CN111062519A (en) | 2020-04-24 |
Family
ID=70298974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911182946.3A Pending CN111062519A (en) | 2019-11-27 | 2019-11-27 | Method and device for sensing customer satisfaction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111062519A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111695819A (en) * | 2020-06-16 | 2020-09-22 | 中国联合网络通信集团有限公司 | Method and device for scheduling seat personnel |
-
2019
- 2019-11-27 CN CN201911182946.3A patent/CN111062519A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111695819A (en) * | 2020-06-16 | 2020-09-22 | 中国联合网络通信集团有限公司 | Method and device for scheduling seat personnel |
| CN111695819B (en) * | 2020-06-16 | 2023-06-02 | 中国联合网络通信集团有限公司 | Seat personnel scheduling method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5907800A (en) | Apparatus for use in inhibiting telecommunication subscriber termination | |
| US8861691B1 (en) | Methods for managing telecommunication service and devices thereof | |
| CN109978608A (en) | The marketing label analysis extracting method and system of target user's portrait | |
| JP2017199377A (en) | System and method for generating subscriber churn prediction | |
| CN110110881A (en) | Power customer requirement forecasting analysis method and system | |
| US20120185540A1 (en) | Method and Arrangement for Supporting Analysis of Social Networks in a Communication Network | |
| CN110580544B (en) | Traffic prediction method based on periodic dependence | |
| CN110248031A (en) | Management method of arranging an order according to class and grade, device, equipment and the storage medium of customer service system | |
| CN104754153A (en) | Customer service system and customer service method thereof | |
| WO2016065885A1 (en) | Alarm information dispatching method and device | |
| CN108632475B (en) | Voice message leaving method and device, computer equipment and storage medium | |
| CN113206909A (en) | Crank call interception method and device | |
| CN111062519A (en) | Method and device for sensing customer satisfaction | |
| CN110490388A (en) | Predict the method and system of user's income | |
| CN110677269B (en) | Method and device for determining communication user relationship and computer readable storage medium | |
| CN113947260A (en) | User satisfaction prediction method and device and electronic equipment | |
| CN105812593A (en) | Method and device for grading users | |
| CN115623130B (en) | Agent conversation service business distribution method and system | |
| US7079634B2 (en) | Apparatus for tracking connection of service provider customers via customer use patterns | |
| EP2919504B1 (en) | Method and device for judging user repeatedly accessing network | |
| CN110782167B (en) | Method, device and storage medium for managing receiving and dispatching area | |
| CN105791613A (en) | Call bill processing method and device | |
| CN110312213B (en) | Operator proportion calculation method and device | |
| KR100751204B1 (en) | System and method for analyzing voice quality per mobile communication service subscriber | |
| CN101621593B (en) | Customer service access control method, device and customer service system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200424 |