TWM629643U - Emotion analysis system - Google Patents

Abstract

An emotion analysis system includes a voice receiver, a server and a processor. The server is configured to store a plurality of voice data. The processor is configured to perform the following steps according to a plurality of commands of the memory: storing a plurality of call record data in the server according to the plurality of voice data; marking an annotation data in the plural call record data according to the plural voice data; obtaining a customer emotion record parameter of the plural call record data according to the annotation data; obtaining a first customer emotion prediction parameter and a first emotion analysis model by applying the plural call record data through the deep learning algorithm; comparing whether the customer emotion record parameter is the same as the first customer emotion prediction parameter; and outputting the first emotion analysis model when the customer emotion record parameter is the same as the first customer emotion prediction parameter.

Description

Sentiment Analysis System

This case is about an analysis system, and in particular a sentiment analysis system.

At present, customer complaints will bring losses to the company (for example, the impact of goodwill, the damage to the brand image, the time and labor cost investment in the aftermath), and may even receive special attention from the competent authorities.

The novel content is intended to provide a simplified summary of the disclosure to give the reader a basic understanding of the disclosure. This novel content is not a complete overview of the present disclosure, and it is not intended to identify important/critical elements of the present embodiments or to delimit the scope of the present disclosure.

One of the technical aspects of this case is about a sentiment analysis system. The sentiment analysis system includes a speech receiver, a server and a processor. The voice receiver is used to collect plural voice data. The server is telecommunicationly coupled to the voice receiver and used for storing plural voice data. The processor is telecommunicationly coupled to the server, and is used for executing the following steps according to the plurality of instructions of the memory: storing the plurality of call record data in the server according to the plurality of voice data; marking the annotation data in the plurality of call record data according to the plurality of voice data; Obtain the customer sentiment record parameters of the multiple call record data according to the annotation data; obtain the first customer sentiment prediction parameter and the first sentiment analysis model by applying the multiple call record data through the deep learning algorithm; compare the customer sentiment record parameters with the first customer Whether the emotion prediction parameters are the same; and when the customer emotion record parameters are the same as the first customer emotion prediction parameters, output the first emotion analysis model.

Therefore, according to the technical content of this case, the emotion analysis system shown in the embodiment of this case can detect customers' emotions in advance, so as to achieve the effect of avoiding customer complaints.

After referring to the following embodiments, those with ordinary knowledge in the technical field of this case can easily understand the basic spirit and other purposes of this case, as well as the technical means and implementation forms adopted in this case.

In order to make the description of the present disclosure more detailed and complete, the following provides an illustrative description for the implementation aspects and specific embodiments of the present case; but this is not the only form of implementing or using the specific embodiments of the present case. The features of various specific embodiments as well as method steps and sequences for constructing and operating these specific embodiments are encompassed in the detailed description. However, other embodiments may also be utilized to achieve the same or equivalent function and sequence of steps.

Unless otherwise defined in this specification, the scientific and technical terms used herein have the same meanings as understood and commonly used by those of ordinary skill in the technical field to which this case belongs. In addition, unless contradicting the context, the singular noun used in this specification covers the plural form of the noun; and the plural noun used also covers the singular form of the noun.

In addition, as used herein, "coupled" may mean that two or more elements are in direct physical or electrical contact with each other, or are in indirect physical or electrical contact with each other, or two or more elements operate with each other or action.

FIG. 1 is a block diagram illustrating a sentiment analysis system according to an embodiment of the present application. As shown, the sentiment analysis system 100 includes a voice receiver 110 , a server 120 and a processor 130 . In connection relationship, the server 120 is telecommunicationly coupled to the voice receiver 110 , and the processor 130 is telecommunicationly coupled to the server 120 .

In order to detect customers' emotions in advance, so as to avoid the occurrence of customer complaints, the present case provides the emotion analysis system 100 as shown in FIG. 1, and the related operations of the emotion analysis system 100 are described in detail below.

Referring to FIG. 1, in one embodiment, the voice receiver 110 is used to collect plural voice data. Then, the server 120 is used for storing the plural voice data. For example, the speech receiver 110 may be a device driven by speech to text (Speech To Text, STT) software or an algorithm, and the speech receiver 110 may capture the words, audio, volume, language spoken by the customer through the speech receiver 110 . Information such as speed, call duration of the entire call, and the ID number entered by the customer in the Interactive Voice Response (IVR), when the customer talks with the customer service staff, the above-mentioned plural voice-related data will be immediately stored in the server The device 120 is used for subsequent analysis, but this case is not limited to this.

Next, the processor 130 is used for executing the following steps according to the plurality of commands of the memory 140 : storing the plurality of call record data in the server 120 according to the plurality of voice data. For example, the memory 140 can be coupled to the processor 130, and the processor 130 can organize plural speech data through data analysis software (such as Python), and segment the words of the client. method (such as the Jieba word segmentation system), using built-in stop words and new words to organize the word data in the plural voice data into call record data, and store it in the server 120, but this case is not limited to this. In addition, the customer's audio, volume, speech rate, and call duration of the entire call can be digitized into call record data, and the call record data can include the customer's ID number and be stored in the server 120 together. The process is to convert unstructured data into structured data, but this case is not limited to this.

Then, the processor 130 is used for marking the annotation data in the plurality of call record data according to the plurality of voice data. Then, the processor 130 is used for obtaining the customer emotion record parameters of the plurality of call record data according to the annotation data. For example, the customer service staff can use the processor 130 to mark the note data in the call record data according to the plurality of voice data. The relevant content specially noted by the customer service staff in the call log data. In addition, the processor 130 can obtain the customer emotion record parameter from the recorded call record data. At this time, the customer emotion record parameter is 1, which means that the customer has a negative emotion. If the call record data is not recorded, the customer emotion record parameter is: 0, but this case is not limited to this.

Then, the processor 130 is used for obtaining the first customer sentiment prediction parameter and the first sentiment analysis model by applying the plurality of call record data through the deep learning algorithm. For example, the deep learning algorithm can be a deep learning model, and the framework of the deep learning model can include Embedding, LSTM and Softmax, and the complex call record data is divided into training set and verification set at a ratio of 8:2, and the The multiple call record data are put into the deep learning model to obtain the first customer sentiment prediction parameters and the first sentiment analysis model, but this case is not limited to this.

Then, the processor 130 is used to compare whether the customer sentiment record parameter is the same as the first customer sentiment prediction parameter. Then, when the customer sentiment record parameter is the same as the first customer sentiment prediction parameter, the processor 130 is used for outputting the first sentiment analysis model. For example, the customer sentiment record parameter may be 0 to 1, and the first customer sentiment prediction parameter may be 0 to 1. When the customer sentiment record parameter is the same as the first customer sentiment prediction parameter, that is, the first sentiment analysis model is used for prediction. The accuracy of customer sentiment is high, and the processor 130 can output and package the first sentiment analysis model into an Application Programming Interface (API), and this API can be used to output customer sentiment prediction parameters, but this case does not apply this limit.

In one embodiment, the processor 130 is further configured to execute the following steps according to the plurality of instructions of the memory 140 : obtaining the second customer emotion prediction parameter through the first emotion analysis model stored in the server 120 according to the plurality of voice data. For example, the plurality of voice data may be the voice content of the customer's call, and the second customer sentiment prediction parameter is obtained through the first sentiment analysis model stored in the server 120 according to the voice content of the customer's call.

FIG. 2 is a diagram illustrating a usage scenario of a sentiment analysis system according to an embodiment of the present application. Please refer to FIG. 1 and FIG. 2 together. In one embodiment, the processor 130 is further configured to execute the following steps according to the plurality of instructions of the memory 140: when the second customer emotion prediction parameter exceeds the emotion parameter threshold, control the The display 150 displays the warning information A1. For example, the emotion analysis system 100 may further include a display 150, the emotion parameter threshold may be 0.8, and the second customer emotion prediction parameter may be 0.9. At this time, since the second customer emotion prediction parameter exceeds the emotion parameter threshold, the warning message A1 is displayed on display 150 . In addition, the second customer sentiment prediction parameter can be called the negative index, and the content of the warning information A1 can be "customer negativeness is too high, please pay attention", but this case is not limited to this.

In other words, the purpose of displaying the warning information A1 on the display 150 is to remind the customer service staff and their immediate supervisors to pay special attention, and immediately take actions to appease the customer to avoid potential customer complaints.

In one embodiment, the processor 130 is further configured to perform the following steps according to the plurality of instructions of the memory 140 : when the customer emotion record parameter is different from the first customer emotion prediction parameter, adjust the hyperparameters in the deep learning algorithm. Then, the processor 130 is used to compare whether the customer emotion record parameter is the same as the third customer emotion prediction parameter. Then, when the customer emotion record parameter is the same as the third customer emotion prediction parameter, the processor 130 is used for outputting the second emotion analysis model. For example, when the customer sentiment record parameters are different from the first customer sentiment prediction parameters, that is, the accuracy of the first sentiment analysis model is too low, so adjust the hyperparameters in the deep learning algorithm until the deep learning algorithm generates The accuracy of the sentiment analysis model is as expected, but not limited to this case.

In one embodiment, the voice data of the plurality of call record data may be word data, data of incoming calls in the day, audio data, volume data, speech rate data or talk time data. For example, the word data can be the word data after the customer's voice is segmented, and the number of incoming calls on the day can be the number of times calculated based on the customer's ID number as a unit. For example, every time the customer contacts the customer service You need to report your ID number once to calculate the number of incoming calls that day. The audio data can be the audio value when the customer contacts the customer service. For example, when the customer is angry to the point that the pitch is broken, the audio data can be 10. The volume data can be the volume value when the customer contacts the customer service. For example, when the customer is angry and roars, the volume data can be 9, and the speech speed data can be the speech speed value when the customer contacts the customer service. For example, When the customer is so anxious that the speech rate is so fast that he can hardly hear it, the speech rate data can be 8, and the call time data can be the total call time when the customer contacts the customer service, but this case is not limited to this.

In one embodiment, the data on the number of incoming calls on the current day is related to the customer ID card data. For example, the number of incoming calls can be the number of incoming calls. The processor 130 can calculate the number of incoming calls in the day from the number of customer ID cards recorded in the server 120 in one day. It is said that if the number of customer ID card data in the server 120 is 3 in one day, it can be calculated that the number of incoming calls on that day is 3, but this case is not limited to this.

It can be seen from the above embodiments of the present case that the application of the present case has the following advantages. The emotion analysis system shown in the embodiment of this case can detect the emotions of customers in advance, so as to avoid the occurrence of customer complaints.

Although the specific examples of this case are disclosed in the above-mentioned embodiments, they are not intended to limit this case. Those with ordinary knowledge in the technical field to which this case belongs can, without departing from the principles and spirit of this case, carry out Various changes and modifications, therefore, the scope of protection in this case should be defined by the scope of the patent application attached hereto.

100: Sentiment Analysis System 110: Voice receiver 120: Server 130: Processor 140: memory 150: Monitor A1: Warning information

In order to make the above and other objects, features, advantages and embodiments of the present case more clearly understood, the descriptions of the accompanying drawings are as follows: FIG. 1 is a block diagram illustrating a sentiment analysis system according to an embodiment of the present application. FIG. 2 is a diagram illustrating a usage scenario of a sentiment analysis system according to an embodiment of the present application. In accordance with common practice, the various features and elements in the figures are not drawn to scale, but are drawn in a manner that best represents the specific features and elements relevant to the present case. Furthermore, the same or similar reference numerals are used to refer to similar elements/components among the different drawings.

100: Sentiment Analysis System

110: Voice receiver

120: Server

130: Processor

140: memory

Claims (10)

A sentiment analysis system that includes: a voice receiver for collecting plural voice data; a server, telecommunicationly coupled to the voice receiver, and used for storing plural voice data; and A processor, telecommunicationly coupled to the server, and configured to execute the following steps according to a plurality of instructions from a memory: to store the plurality of call log data in the server according to the voice data; marking a note data on the call log data according to the voice data; to obtain a customer sentiment record parameter of the call record data according to the registered data; Passing the call record data through a deep learning algorithm to obtain a first customer sentiment prediction parameter and a first sentiment analysis model; comparing whether the customer sentiment record parameter is the same as the first customer sentiment prediction parameter; and When the customer sentiment record parameter is the same as the first customer sentiment prediction parameter, the first sentiment analysis model is output. The emotion analysis system of claim 1, wherein the processor is further configured to perform the following steps according to the instructions of the memory: According to the voice data, a second customer emotion prediction parameter is obtained through the first emotion analysis model stored in the server. The emotion analysis system of claim 2, wherein the processor is further configured to perform the following steps according to the instructions of the memory: When the second customer emotion prediction parameter exceeds an emotion parameter threshold, a display is controlled to display a warning message. The emotion analysis system of claim 2, wherein the processor is further configured to perform the following steps according to the instructions of the memory: When the customer emotion record parameter is different from the first customer emotion prediction parameter, a hyperparameter in the deep learning algorithm is adjusted. The emotion analysis system of claim 4, wherein the processor is further configured to perform the following steps according to the instructions of the memory: A third customer sentiment prediction parameter and a second sentiment analysis model are obtained from the voice data through the deep learning algorithm. The emotion analysis system of claim 5, wherein the processor is further configured to perform the following steps according to the instructions of the memory: Compare whether the customer sentiment record parameter is the same as the third customer sentiment prediction parameter. The emotion analysis system of claim 6, wherein the processor is further configured to execute the following steps according to the instructions of the memory: When the customer sentiment record parameter is the same as the third customer new sentiment prediction parameter, the second sentiment analysis model is output. The emotion analysis system according to claim 1, wherein the call record data includes word data, incoming phone number data, audio data, volume data, speech rate data, and talk time data at least one of them. The emotion analysis system of claim 1, wherein the call record data includes a client ID card data. The sentiment analysis system according to claim 9, wherein the data on the number of incoming calls on a day is related to the customer ID card data.

Images