JP7445932B2 - Information processing device, information processing method and program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a program for evaluating communication via a terminal.

Partly due to the influence of the coronavirus, in recent years, sales have increasingly been conducted via applications such as LINE (registered trademark). On the other hand, unlike face-to-face sales, sales through applications can make it difficult to get a sense of distance from customers, and some salespeople may not know how to communicate with customers. . They often don't know whether they are communicating with their customers properly or not, and they often don't know how to improve their sales activities.

Although proposals for sales support have been made in the past (for example, Patent Documents 1 and 2), no means for dealing with the above problems have been proposed yet.

JP 2022-002059 Publication JP2021-068036A

The present invention was made based on the above-mentioned problem, and provides an information processing device and the like that can evaluate communication via a terminal. Although the origin of the present invention is a problem related to business, the field to which the present invention is applied is not limited to business.

The information processing device according to the present invention includes:
an acquisition unit that acquires communication information regarding communication via the terminal;
Evaluate the characters included in the communication after performing natural language processing, evaluate the number of times or number of people who communicated within a given period, or evaluate the time taken to reply to a transmission from a customer in communication information or the response to a transmission to a customer. a first evaluation department that evaluates the time it takes to receive a response from a customer;
may be provided.

The information processing device according to the present invention includes:
It may also include an output section that outputs information based on the evaluation results of the first evaluation section for a plurality of people.

In the information processing device according to the present invention,
The acquisition unit acquires excellent communication information via the terminal of one or more predetermined excellent people who have received excellent evaluations,
The first evaluation department evaluates excellent communication information after performing natural language processing on the characters included in the communication, evaluates the number of times or number of people who communicated within a predetermined period, or even responds to customer transmissions. or the time it takes for a customer to respond to a message sent to a customer;
The output unit may output the evaluation result of the excellent communication information by the first evaluation unit.

In the information processing device according to the present invention,
The acquisition unit acquires general communication information via the terminal of one or more ordinary people different from the excellent person,
The first evaluation department evaluates the general communication information after performing natural language processing on the characters included in the communication, evaluates the number of times or number of people who communicated within a predetermined period, or even responds to customer transmissions. or the time it takes for a customer to respond to a send to a customer;
The output unit may output the evaluation result of the general communication information by the first evaluation unit.

In the information processing device according to the present invention,
The first evaluation department evaluates using the number of times or percentage of use of a trend word over a certain period of time, or the percentage of replies with pictograms or stamps to messages sent by customers with pictograms or stamps included in communication information. It's okay.

In the information processing device according to the present invention,
The first evaluation unit may perform evaluation using the number of appearances or the appearance rate of a keyword in a certain period of time, which is included in the communication information and indicates gratitude, joy, or request from the customer.

In the information processing device according to the present invention,
The first evaluation unit gives a high evaluation when the average time to a reply to a transmission from a customer in communication information or the average time to a reply from a customer to a transmission to a customer in communication information is shorter than a threshold; If it is longer than the threshold, it may be given a low evaluation.

In the information processing device according to the present invention,
The first evaluation unit determines keywords indicating compliance violations included in the communication information on the terminal,
If a keyword indicating compliance violation is included, a warning may be sent to the terminal performing the communication.

In the information processing device according to the present invention,
It may also include a notification section that recommends communication with the customer based on the customer information.

In the information processing device according to the present invention,
A second evaluation unit may be provided that evaluates the disclosure level of personal information disclosed by a customer and gives a high evaluation to a customer with a high disclosure level.

The information processing method according to the present invention includes:
a step of acquiring communication information regarding interaction via the terminal by the acquisition unit;
The first evaluation department evaluates the characters included in the communication after performing natural language processing, evaluates the number of times or number of people who communicated within a predetermined period, or evaluates the time required to reply to a customer's transmission of communication information. or evaluating the time it takes for a customer to respond to a message sent to the customer;
may be provided.

The program according to the present invention is
A program installed on an information processing device,
The information processing device in which the program is installed is
A function to obtain communication information regarding interactions via the terminal,
Evaluate the characters included in the communication after performing natural language processing, evaluate the number of times or number of people who communicated within a given period, or evaluate the time taken to reply to a transmission from a customer in communication information or the response to a transmission to a customer. Ability to evaluate the time it takes for a customer to respond,
may be provided.

According to the present invention, it is possible to provide an information processing device and the like that can evaluate communication via a terminal.

FIG. 1 is a block diagram of an information processing system according to an embodiment of the present invention. FIG. 2 is a diagram showing how the results of data analysis are visualized in the embodiment of the present invention. FIG. 3 is a diagram for explaining services provided by introducing the information processing system according to the embodiment of the present invention, such as visualization, recommendations, and discovery of potential customers. FIG. 4 is a diagram for explaining the manner in which the amount of sales activities and the like are visualized in the embodiment of the present invention. FIG. 5 is a diagram for explaining a manner in which a reminder is notified in the embodiment of the present invention. FIG. 6 is a diagram for explaining a mode for displaying highly accurate persons in the embodiment of the present invention. FIG. 7 is a diagram for explaining a usage mode regarding compliance violation detected by the information processing apparatus according to the embodiment of the present invention.

Embodiment 《Configuration》
Embodiments of an information processing system and an information processing device 100 according to the present invention will be described below. In this embodiment, an information processing method using the information processing device 100, a program installed to generate the information processing device 100, and a storage medium such as a USB or DVD that stores the program are also described in this embodiment. Provided by. Additionally, programs that are installed on various terminals 200 such as personal computers, smartphones, and tablets are also provided. In this embodiment, "or" is a concept that includes "and", and A or B means either A, B, or both A and B.

The information processing system of this embodiment may include an information processing device 100 and one or more terminals 200. Information processing device 100 is typically a server, and may be placed in a cloud environment.

As shown in FIG. 1, the information processing apparatus 100 according to the present embodiment includes an acquisition unit 10 that acquires communication information regarding, for example, interactions with customers via the terminals 200 from a plurality of terminals 200, and A first evaluation unit 130 functions as a communication evaluation unit that evaluates a terminal holder such as a sales person who owns the terminal 200 after performing natural language processing on the characters, and the first evaluation unit 130 outputs the evaluation results. , and an output unit 70 that allows the evaluation results by the first evaluation unit 130 to be viewed on a plurality of terminals 200. The first evaluation unit 130 may evaluate the communication style of the terminal owner using pictograms, stamps, etc. included in the communication information. The first evaluation unit 130 may evaluate the communication style of the terminal owner based on information regarding the number of times or the number of people who communicated. The first evaluation unit 130 is capable of distinguishing between communications issued by a person in charge who owns the terminal 200 and communications issued by a customer or the like with whom the communication is being conducted. The "customer" in this embodiment also includes potential customers. Interactions with customers using the terminal 200 may be performed via a general application such as LINE (registered trademark) or a dedicated application. Information input from the terminal 200, including exchanges via applications, may be performed via the information processing device 100 consisting of a server or the like, or may be transmitted from the terminal 200 to the information processing device 100 at any time or periodically. Good too. Exchanges between the customer and the person in charge via the terminal 200 may be stored in the storage unit 60. When communication between a customer and a person in charge is performed via an information processing device 100 consisting of a server or the like, the acquisition unit 10 may acquire communication information through communication written directly in the information processing device 100. . The interaction with the customer via the terminal 200 may be stored in the storage unit 60 of the information processing device 100 in association with the identification information of the terminal 200 and the identification information of the person in charge, and may also be similarly linked with the identification information of the customer. The data may then be stored in the storage unit 60 of the information processing device 100.

The terminal 200 used in this embodiment may be a smartphone, a tablet terminal, a personal computer, or the like. When the terminal 200 is a smartphone or a tablet terminal, an input display section consisting of a touch panel that also serves as an input section 220 and a display section 210 is provided. Information output from the output unit 70 can be displayed on the terminal 200. The terminal 200 of this embodiment is a concept that includes a management terminal 200a used by an administrator and a general terminal used by a person other than the administrator. In this embodiment, the input section of the management terminal 200a is shown as 220a, and the display section is shown as 210a.

The communication information evaluated by the first evaluation unit 130 may include the number of customers with whom communication has been performed within a predetermined period, the number of messages to customers, or the number of interactions. By evaluating these items by the first evaluation unit 130, it is possible to evaluate whether active communication is being attempted with customers, and it is possible to evaluate the amount of sales activities of each person in charge (for example, a sales person). The predetermined period may be set as appropriate, and may be one day, one week, or one month. The predetermined period may be determined in advance, or may be specified from the management terminal 200a. The first evaluation unit 130 may give a high evaluation when the number of customers with whom the customer has interacted is large within a predetermined period, when the number of messages to customers is large, when the number of interactions is large, etc. The first evaluation unit 130 may evaluate the number of customers with whom it interacts, and the higher the number of customers it interacts with, the higher the evaluation may be. The number of customers served may be the number of customers with whom the customer has communicated via LINE or the like within a predetermined period. As mentioned above, interactions with customers are linked to the customer's identification information and the identification information of the person in charge or the terminal 200 owned by the person in charge, and it is possible to track which customer and which person in charge interact with each other and how often. The first evaluation unit 130 may be able to perform the evaluation. When communication information is stored in the storage unit 60 in association with identification information such as an identification number or identification symbol for identifying a person in charge such as a sales person, it is important to note that the communication information has been communicated with each person in charge within a predetermined period. The number of customers, the number of messages sent to customers, or the number of exchanges will be stored in the storage unit 60.

The threshold values for "number" and "ratio" in this embodiment may be stored in advance in storage unit 60, or may be input from input unit 220a of management terminal 200a. You may evaluate whether the number is large or small, or whether the percentage is high or low, based on a threshold number or percentage.

The first evaluation unit 130 may also evaluate changes in the amount of sales activities of each person in charge (for example, a sales person). By looking at changes in the amount of sales activity (by observing conversation situations over a long period of time), it is possible to measure whether there has been a recent change in each person's desire to work or whether there have been ups and downs in motivation. Additionally, by looking at changes in the amount of sales activity over a longer period of time, it is also possible to observe changes in the sales style of each person in charge, and measure the growth of each person in conjunction with individual performance status. . In addition, by evaluating the communication information by the first evaluation unit 130, when a problem occurs, the first evaluation unit 130 can evaluate the time taken to solve the problem, the number of chats, etc. from the communication information, and the person in charge It is also possible to evaluate one's work ability as a person (for example, a salesperson). In the case of an excellent amount of sales activities, a symbol or display may be displayed on the display unit 210 of the terminal 200 to commend the results.

As described above, not only the person in charge but also the customer has identification information such as an identification number and an identification symbol, and the customer's information may be managed using the customer's identification information. In this case, communication information with the customer may be stored in the storage unit 60 in association with the customer's identification information. By linking business-related topics that often come up in conversations with customers with customer information and storing them in the storage unit 60, the needs of each customer can be grasped and predictive sales become possible. Further, the first evaluation unit 130 may evaluate the needs of a large number of customers in an integrated manner, and in this case, it becomes possible to grasp recent general needs trends. Evaluation may be performed for each customer category, in which case it is also possible to grasp trends in customer needs for each category.

The amount of sales activities, such as the actual use of SNS, including the actual results of conversations on LINE, etc., of multiple representatives may be visualized in a list and displayed on a shared screen in the sales office, etc. (see FIGS. 2 to 4). The plurality of target personnel may be sales personnel at the same sales office, sales personnel in the same sales area, or sales personnel nationwide. The target range may be input as appropriate from the input unit 220a of the management terminal 200a.

The output unit 70 may visualize the amount of sales activity by outputting information based on the evaluation results of the first evaluation unit 130 for a plurality of terminal holders (for example, a plurality of sales representatives). When visualizing the amount of sales activities in this way, the average value of the number of customers who have talked, the number of times they have talked, the number of times they have talked per customer, etc. may be displayed. By adopting such an aspect, it can be expected that, for example, a sense of competition within the business office will be maintained and improved. Also, by looking at the graph posted, you can reaffirm your position and the situation around you. In addition, the output unit 70 outputs the number of customers, the number of conversations, the number of conversations per customer, the average number of customers, the number of conversations, the number of conversations per customer, etc. of the top customers with a high usage record, and sales It may also be possible to display the information on a shared screen at a certain location. Further, the notification unit 30 may notify the terminal 200 of each sales person, etc., by push notification or the like, so that the information can be confirmed on the terminal 200 owned by the person in charge. In this case, it is possible to understand the points you are lacking in comparison with several top people with a high usage record and the comparison with the average value, which can be used as a trigger for behavior improvement.

In the natural language processing performed by the first evaluation unit 130, characters included in the communication may be divided into words by morphological analysis. The first evaluation unit 130 may also perform syntactic analysis, semantic analysis, context analysis, dependency analysis, etc. on the character string. The first evaluation unit 130 has an artificial intelligence function and creates a feature vector corresponding to each item based on the proximity between words used in a large amount of text data (for example, words used on the web). It's okay. The first evaluation unit 130 may then use the feature vector to calculate the degree of similarity between the words used in communication and each item. The proximity of words is estimated and determined by the artificial intelligence function of the first evaluation unit 130, for example, from the way the words are used (context of the words, etc.) on sites on the web (including SNS sites, etc.). (learned) and may be updated from time to time or as appropriate. Information regarding the proximity of words may be stored in the storage unit 60.

Further, as an example of the artificial intelligence function, a classifier using a machine learning method may be used. This classifier may, for example, use machine learning technology to determine adopted variables (elements) and their coefficients (weights) to be used, so as to generate highly accurate vectors from past performance data. The artificial intelligence function may perform regression analysis, decision tree analysis, etc. Regarding machine learning techniques, various models can be employed, such as a logistics regression model, a random forest model, a tree model, etc.

Machine learning may be performed by dividing past actual communication information into learning data and test data. As an example, communication information for a large number of personnel can be divided into learning data and test data, and the learning data can be used to give high marks to the communication information of employees with good sales performance, and to give high marks to the communication information of employees with poor sales performance. is given a low evaluation, and the communication information of an employee whose sales performance is at an average level is given a medium evaluation, and the first evaluation section 130 performs machine learning. Then, the accuracy of the evaluation model generated by machine learning using the test data is confirmed. At this time, the learning data and test data may be replaced using a cross-validation method. Machine learning uses information such as the number of customers with whom you have interacted within a given period of time, the number of messages or interactions with customers, the number of times or percentage of use of trend words in a certain period of time, and the number of emojis or stamps from customers included in communication information. Input information such as the percentage of replies with emoticons or stamps in response to transmissions, the number of times or percentage of keywords that indicate gratitude, joy, or request from customers included in communication information appear in a certain period of time, etc. An evaluation may also be performed. The evaluation model may be prepared in advance, input from the management terminal 200a, and stored in the storage unit 60.

The first evaluation unit 130 may mine the proximity between words used on the web. The first evaluation unit 130 picks up a plurality of words related to each item from the Big Text Data (creates a classified vocabulary list), and also Then, a feature vector associated with each item may be created.

The first evaluation unit 130 may use the feature vectors created in this way to interpret the meanings of terms included in the communication information. The importance level of a word such as "particle" may be lowered (for example, it may be set to "0"). In addition to vectorization, the first evaluation unit 130 may perform clustering and association strength analysis using machine learning.

The first evaluation unit 130 evaluates the number of times or percentage of use of a trend word in a certain period of time, or the percentage of replies with pictograms or stamps to the transmission of pictograms or stamps from customers included in the communication information. Good too. By evaluating these items by the first evaluation unit 130, it is possible to evaluate the behavioral patterns regarding speaking skills of each person in charge (for example, a sales person). The first evaluation unit 130 gives a high evaluation when the number of uses of the trend word in a certain period is large or the use ratio is higher than a predetermined threshold, and the first evaluation unit 130 gives a high evaluation when the number of uses of the trend word in a certain period is low or the use ratio is higher than a predetermined threshold. A low evaluation may be given when the value is lower than a predetermined threshold. The first evaluation unit 130 gives a high evaluation when the percentage of replies with pictograms or stamps to the transmission of pictograms or stamps from the customer is higher than a predetermined threshold, and A low evaluation may be given if the percentage of replies with pictograms or stamps in response to transmissions is lower than a predetermined threshold. You can evaluate whether you are communicating with customers by incorporating current events based on the number of times and percentage of use of trending words over a certain period of time. The first evaluation unit 130 evaluates the percentage of customers who reply with stamps to stamps, and the percentage of customers who reply with stamps, pictograms, or emoticons to pictograms or emoticons from customers. You can evaluate whether stamps, emoticons, and emoticons are used appropriately according to characteristics. The search unit 20 may automatically search for trend words on the WEB, SNS, etc., or the administrator may input them as appropriate from the input unit 220a of the management terminal 200a. The trend word may be stored in the storage unit 60. The search unit 20 may update the trend words stored in the storage unit 60. As trend words, recent topics that can be searched on the web (for example, Big Boss, Russia, virtual currency, etc.) may be adopted. In this case, the first evaluation unit 130 It may also be possible to evaluate how many recent topic words appear.

The first evaluation unit 130 may evaluate the number of appearances or the appearance rate of a keyword indicating gratitude, joy, or request from a customer, which is included in the communication information between the person in charge and the customer, in a certain period of time. This information is also included in behavior patterns related to speaking skills, and it is possible to evaluate the degree to which the person in charge is trusted by the customer. If the number of appearances of a keyword indicating gratitude, joy, or request from a customer is greater than a predetermined threshold, or if the appearance rate is higher than a predetermined threshold, the first evaluation unit 130 You may also give a high evaluation. On the other hand, if the number of occurrences of keywords indicating gratitude, joy, or requests from customers is lower than a predetermined threshold, or if the appearance rate is lower than a predetermined threshold, the first evaluation unit 130 You may give a low rating to the information. The first evaluation unit 130 may evaluate whether the keyword corresponds to a keyword indicating gratitude, joy, or request from the customer using information stored in the storage unit 60. When the first evaluation unit 130 has an artificial intelligence function and performs machine learning, as a result of the learning, it determines whether the input information from the other party included in the communication information corresponds to keywords related to gratitude, joy, or asking for a favor. You may. The keywords indicating gratitude, joy, and requests may be updated to the storage unit 60 by the administrator via the management terminal 200a.

By the first evaluation unit 130 evaluating the communication information for each person in charge, the notification unit 30 may provide advice to the terminal 200 of each person in charge regarding how to communicate. For example, the first evaluation unit 130 may perform a log analysis and may indicate that it is preferable to use polite language for customers in their 30s, or encourage female customers in their 40s to use popular stamps. Such advice information may be input from the management terminal 200a and stored in the storage unit 60, or information stored in the storage unit 60 as a result of analyzing communication information of excellent people, which will be described later, may be used. .

The first evaluation unit 130 may evaluate the average time until a reply to a transmission from a customer in communication information or the average time until a reply from a customer to a transmission to a customer. This evaluation item is information included in behavior patterns related to time consciousness, and by performing such evaluation, it becomes possible to understand the best response time of customers and contact them. The time required for the response may be calculated by the first evaluation unit 130 character-recognizing the time displayed as the transmission time and the time displayed in the reply to the transmission. For example, the difference between the transmission time from the person in charge and the response time from the customer is the time until the customer replies to the transmission to the customer. Further, when communication information is transmitted via the information processing device 100, information regarding each of the transmission time and the reply time is stored as is in the storage unit 60, and as a result, the time required to send a reply to a customer's transmission is The storage unit 60 stores the time and the time until the customer replies to the transmission to the customer. Furthermore, each time communication is performed, a notification to that effect is transmitted from the terminal 200 to the information processing device 100, and as a result, the time taken to reply to a transmission from a customer or the time taken until a reply from a customer to a transmission to a customer is 60 may be stored. If the average time to reply to a transmission from a customer is shorter than a predetermined threshold, the first evaluation unit 130 gives a high evaluation to the person in charge, and waits until the reply from the customer to the transmission to the customer. If the average time is shorter than a predetermined threshold, the person in charge may be highly evaluated. If the average time to reply to a transmission from a customer is longer than a predetermined threshold, the first evaluation unit 130 gives a low evaluation to the person in charge, and waits until the customer replies to the transmission to the customer. If the average time is longer than a predetermined threshold, the person in charge may be given a low evaluation. If the average time it takes to reply from a customer is short, there is a high possibility that the customer is convenient for that time period, so the storage unit stores the convenient time period for each customer by associating it with the customer's identification information. 60 may be stored. The average time may be an average time in a predetermined period, or may be an accumulated average time from the past to the present.

The notification unit 30 may recommend that the customer in charge communicate with the customer based on the customer information of the person in charge linked to the terminal 200 (see FIGS. 3 and 5). Examples of customer information used for recommendations include the customer's birthday, the birthdays of the customer's family, and anniversary information such as the customer's wedding anniversary. You may also recommend communicating with potential customers you have not approached in the recent past (for example, within a month). If such a mode is adopted, it will be possible to understand anniversaries and customers who have not been approached recently, which can be expected to lead to an increase in the frequency of approaches. It is also expected that efficiency can be improved by using standard greetings after participating in an event or on anniversaries. The notification unit 30 may also provide a reminder even if the customer has not responded to a message despite receiving a message from the customer.

The timing of communicating with customers may be customized by each sales staff member inputting information from the input unit 220 of his or her own terminal 200. Further, the recommended timing for communicating with customers may be before the so-called third time period, such as during lunch break or from 7:00 pm to 9:00 pm on weekdays. The notification may be sent to the terminal 200.

The notification unit 30 measures and notifies customers who have a short sales approach period, such as those who participated in an event or those who were introduced by a customer, as to whether or not they will become your definite customers. Information about the customer may be displayed on the terminal 200 of the person in charge of the customer (see FIGS. 3 and 6). The degree of accuracy may be stored in the storage unit 60 together with the customer's identification information. The “accuracy” stored in the storage unit 60 may be analyzed by the second evaluation unit 140. The second evaluation unit 140 evaluates the probability of becoming a customer using information such as the disclosure level described later, the time required to reply to the transmission from the person in charge (speed of response), and whether the inquiry is from a good customer. It may be calculated. The second evaluation unit 140 may also perform machine learning to generate the accuracy model. In this case, information regarding multiple customers, such as past disclosure level, time required to reply to a transmission from a representative (speed of response), and whether the inquiry is from a good customer, is combined into learning data and test data. The second evaluation unit 140 may perform machine learning to give a high evaluation if the customer actually becomes a customer, and a low evaluation if the customer does not become a customer. At this time, the learning data and test data may be replaced using a cross-validation method. In this embodiment, a "determined customer" refers to a customer who has actually purchased a product promoted by a salesperson or received a service promoted by a salesperson. By adopting such an aspect, it becomes possible to know who is likely to become a definite customer, and it is possible to efficiently approach potential customers.

The storage unit 60 may store the disclosure level of personal information disclosed by the customer for each customer, and the second evaluation unit 140 may give a high evaluation to a customer with a high disclosure level. The higher the disclosure level, that is, the closer the disclosure level is to personally identifiable information, the higher the second evaluation unit 140 may give the customer a higher evaluation. For example, level 0 (0 points) is a customer who only discloses a nickname, level 1 (1 point) is a customer who discloses their real name, contact information other than SNS (phone number, email address), and level 2. (2 points) are customers who disclose their address, level 3 (3 points) are customers who disclose their anniversaries (birthdays, wedding anniversaries), and level 4 (4 points) are customers who disclose family information. I can name a customer who does. The disclosure level of personal information may be ascertained by a person in charge inputting customer information from the terminal 200, or may be ascertained by natural language processing of communication information by the first evaluation unit 130.

The first evaluation unit 130 may evaluate a keyword indicating a compliance violation, which is included in the communication information in the terminal 200. If a keyword indicating a compliance violation is included, the notification unit 30 may notify a warning to the terminal 200 of the person in charge who is communicating. By adopting such an aspect, it is possible to make the customer take appropriate actions while making the customer comply with the rules. If the number of message transmissions including a keyword indicating compliance violation is greater than a predetermined threshold or the percentage is high, the first evaluation unit 130 may give a low evaluation. Examples of compliance violations include disclosure of other customers' personal information, disclosure of contract numbers, requests for expensive gifts, and power harassment.

The first evaluation unit 130 may analyze the number of transmissions or the transmission ratio of keywords indicating compliance violation in a certain period of time. When a term corresponding to a compliance violation is stored in the storage unit 60 and a term corresponding to a compliance violation or a term similar to the term is used in the communication information, the notification unit 30 sends a message to the terminal 200 of the person in charge who is conducting the communication. A warning may be issued. A term indicating a compliance violation may be stored in the storage unit 60. The administrator may update such terms via the management terminal 200a.

When a term indicating a compliance violation is input on the terminal 200, the notification section 30 that has received the analysis result by the first evaluation section 130 may transmit warning information such as an alert to the terminal 200.

If a term indicating a compliance violation is detected, an alarm may be displayed to the person in charge and the person in charge. For example, when a customer sends a term requesting the disclosure of personal information, an alarm may be displayed on the terminal 200 of the person in charge to warn the person to be careful in responding. By displaying an alarm on the terminal 200 in this way, it is possible to reduce the level of concern that the person in charge has to worry about whether or not a compliance violation has occurred, and it is also expected to reduce stress on the person in charge. Further, by displaying an alarm on the terminal 200, it is possible to prevent the person in charge from providing inappropriate information. Furthermore, the content of the alarm display may also be displayed on the terminal 200 of the person in charge.If such a mode is adopted, a prompt response from the person in charge can be expected, and the person in charge can also be informed. It is also expected that the person in charge will respond more appropriately because it assumes that there is a person in charge. This will also lead to improved services for customers. Log analysis showing keywords indicating compliance violations can also be reflected in training texts for personnel (see Figure 7). It will be possible to create case studies, create countermeasure plans based on high-frequency words, and conduct training and countermeasure training.

When visualizing the amount of sales activities, a first creation unit 110 that creates visualization information such as a graph or a table based on information included in the communication information may be provided. The first creation unit 110 generates visualization information such as graphs or tables for each item such as number of customers handled, number of messages handled, trend keyword usage rate, stamp usage rate, emoticon usage rate, customer satisfaction, message sending timing, compliance violation rate, etc. You may also create one. Customer satisfaction may be indicated by an evaluation result based on the number or percentage of gratitude, joy, or requests included in the communication information. The evaluation by the first evaluation unit 130 based on the speed of response from customers may be shown by an evaluation result based on the number or percentage of responses received within a predetermined time (for example, one hour), for example. The predetermined response time may be set in multiple stages, with a high rating (3 points) for a response within the first time, a medium rating (2 points) for a response within the second time exceeding the first time, and a medium rating (2 points) for a response within the second time. It is also possible to give a low rating (1 point) if the test exceeds the third hour, and no rating (0 points) if the test exceeds the third time.

A second creation unit 120 may be provided that creates sentences to explain the visualized information such as graphs or tables created by the first creation unit 110. The second creation unit 120 automatically creates an explanatory text by analyzing the content of visualization information such as a graph or a table.

The second creation unit 120 may create sentences using LSTM (Long Short Term Memory). LSTM is classified as an RNN (recurrent neural network) among neural networks. RNN is a neural network for learning time-series data, and LSTM is an RNN that has been repeatedly improved. By using LSTM, time-series data can be taken into consideration, and learning can be performed taking into account the order of word strings. Tensorflow, a software library for machine learning, may also be used. As the AI model, a plurality of Deep Learning (Seq2Seq type models) may be used. Beam search may be used when performing machine learning, or reinforcement learning may be performed in which a penalty is applied when there is a mismatch between input data and output text.

The output unit 70 may output the visualization information such as a graph or a table created by the first creation unit 110 and the explanatory text created by the second creation unit 120 together.

Next, a mode using excellent communication information of an excellent person will be explained.

The acquisition unit 10 may acquire excellent communication information regarding one or more predetermined excellent people who have received excellent evaluations and their interactions with customers in text via the terminal 200. The first evaluation unit 130 may analyze the amount of sales activities or behavior patterns included in the excellent communication information. The analysis result may be output by the output unit 70 to, for example, the storage unit 60 and stored in the storage unit 60. Further, the analysis results by the first evaluation unit 130 are outputted by the output unit 70 to the display unit 210 such as the display screen of the terminal 200 of each person in charge, and the analysis results are displayed on the display unit 210. good.

The acquisition unit 10 may acquire general communication information about one or more ordinary people who are different from the excellent people, regarding interactions with customers in text via the terminal 200. The first evaluation unit 130 may analyze the amount of sales activities or behavior patterns included in the general communication information. A comparison unit 40 may be provided to compare the analysis results of excellent communication information and the analysis results of general communication information. In this case, the output unit 70 may output the comparison result by the comparison unit 40. An excellent person is, for example, an employee or staff member who has achieved excellent sales performance. Persons, on the other hand, are other employees and staff. Typically, there are multiple excellent people and multiple ordinary people, but the number of excellent people is smaller than the number of ordinary people.

For excellent communication information when there are multiple excellent persons, the average result of the analysis results of each excellent person is used. Similarly, for general communication information when there are a plurality of general persons, the result of averaging the analysis results of each general person is used. The identification of outstanding individuals may be carried out by the administrator via the management terminal 200a, or the selection unit may automatically identify individuals who have achieved performance above the first threshold based on the sales performance etc. stored in the storage unit 60. The selection may be made by By performing analysis using excellent communication information, it will be possible to analyze and compare differences between excellent people and omissions, biases, excesses and deficiencies in information, and it will also be possible to realize automatic education.

The person in charge may input his or her own identification information and then be able to obtain the analysis results for his or her own communication information. In this case, in addition to the input of identification information, the input of a password is requested, and when it is confirmed that these information are correct, the output unit 70 outputs the analysis result of the communication information of the person in charge. You may also do so. In this case, the acquisition unit 10 acquires the target communication information regarding the text communication with the customer via the terminal 200 of the person in charge, and the amount of sales activities included in the target communication information evaluated by the first evaluation unit 130 or A comparison result between the behavior pattern, the analysis result of the excellent communication information, and the analysis result of the target communication information may be displayed. Note that the management terminal 200a may be configured to allow the analysis results of the communication information of each person in charge to be confirmed as appropriate. Access to the management terminal 200a may also require input of a password in addition to input of identification information.

It may also be possible to specify the period of analysis results to be output. For example, when a target period (for example, one month, etc.) is input in the input unit 220, the analysis results for the target period are read from the storage unit 60 and output by the output unit 70.

The first evaluation unit 130 may also analyze the number of customers registered in the terminals 200 of excellent people, target people (for example, individual sales representatives), and ordinary people (for example, other sales representatives) as the amount of sales activities. good. The first evaluation unit 130 may analyze the average number of interactions of excellent persons, target persons, and general persons with individual customers over a certain period of time as the amount of sales activity. The target person may be recognized by the control unit 50 based on identification information requested at the time of login. The fixed period is a predetermined period such as one day, one week, one month, etc., and may be set in advance or may be specified from the input section 220. As an example, the average number of interactions with one customer via the terminal 200 in a day may be output by the first evaluation unit 130 as the amount of sales activity.

The first evaluation unit 130 may analyze the number of times or usage rate of trend words included in the excellent communication information, target communication information, and general communication information in a certain period of time as a speaking style included in the behavior pattern.

The first evaluation unit 130 calculates the percentage of replies with emoticons or stamps to the transmission of emoticons or stamps from customers included in the excellent communication information, target communication information, and general communication information, as a speaking skill included in the behavior pattern. You can analyze it. The first evaluation unit 130 may analyze the ratio of replies using emoticons to transmissions using emoticons from customers, or the ratio of replies using stamps to transmissions using stamps from customers. It's okay. The first evaluation unit 130 may analyze the number of times or percentage of use of pictograms or stamps in a certain period of time, which are included in the excellent communication information, target communication information, and general communication information.

The display unit 210 may display a comparison result of an analysis result by an excellent person, an analysis result by a target person, and an analysis result by an ordinary person. The display mode may be divided by item. For example, when the number of customers served is selected, the number of customers served by excellent people, target people, and general people may be compared and displayed using a graph such as a bar graph. The same applies if you select other items, such as number of messages supported, trend keyword usage rate, stamp usage rate, emoji usage rate, customer satisfaction, message sending timing, compliance violation rate, etc. The analysis results for excellent people, target people, and ordinary people are displayed on the display unit 210 for each item.

In the dashboard shown in FIG. 2, the number of customers handled, the number of messages handled, trend keyword usage rate, stamp usage rate, emoji usage rate, Customer satisfaction, message sending timing, and compliance violation rate can be displayed. For example, if you tap ``Number of Customers Corresponded'', the graph shown in the lower right corner of Figure 2 will be displayed, allowing you to visually check the ``Number of Customers Corresponded'' for Excellent Persons, Target Persons, and General Persons. There is.

Only the information regarding the excellent person may be analyzed by the first evaluation unit 130 and outputted by the output unit 70 without being compared by the comparison unit 40. Furthermore, the information regarding excellent people and the information regarding ordinary people may be analyzed by the first evaluation unit 130, compared by the comparison unit 40, and the results may be output. Further, the information regarding the excellent person and the information regarding the target person may be analyzed by the first evaluation section 130, compared by the comparison section 40, and the results may be output.

The types and periods of excellent communication information, target communication information, and general communication information read by the acquisition unit 10 may be input and specified from the terminal 200. The same period may be automatically specified for each of the excellent communication information, target communication information, and general communication information. However, the present invention is not limited to this embodiment, and the periods for each of excellent communication information, target communication information, and general communication information may be set to be different.

Obtain excellent communication information regarding text interactions with customers via the terminal 200 of one or more predetermined excellent persons who have received excellent evaluations, and analyze the amount of sales activities or behavioral patterns included in the excellent communication information. If this mode is adopted, it will be possible to show the analysis results of the interactions with customers using text (including pictograms and stamps) via the terminal 200 such as a smartphone of an excellent person who has received an excellent evaluation. become. Therefore, it becomes possible to refer to the approaches to customers of excellent people, such as people with excellent sales skills.

The acquisition unit 10 acquires general communication information regarding one or more ordinary people different from excellent people's interactions with customers in text via the terminal 200, and the first evaluation unit 130 acquires sales activities included in the general communication information. If the amount or behavior pattern is analyzed and the comparison unit 40 compares the analysis results of excellent communication information with the analysis results of general communication information, the customer of an excellent person, such as a person with excellent sales ability, It is possible to compare this approach with that of an ordinary person who does not have very good sales skills, and as a result, it becomes possible to see where the differences between the two arise. For example, the items that differentiate the two may differ depending on the type of industry (frequency of emoticons, number of interactions, etc.), but by adopting this method, it is possible to determine where the differences between excellent people and ordinary people emerge. can be easily checked.

Note that ordinary people may be divided into a first ordinary person who has an average performance, and a second ordinary person who has a performance considerably lower than the average. In this case, the analysis results for the normal person and the comparison results based on the analysis results can be shown separately for the first normal person and the second normal person. Therefore, it is expected that it will be possible to analyze the factors as to why the second normal person's performance is much lower than the average.

The second normal person may be specified by the administrator via the management terminal 200a, or the selection unit selects a person whose performance is less than the second threshold from the sales results etc. stored in the storage unit 60. may be selected automatically.

When the first evaluation unit 130 analyzes the amount of sales activities based on the number of customers registered in the terminals 200 used by outstanding people and ordinary people, it is important to consider how many people are registered as customers. You will be able to analyze and make comparisons based on this information. As for whether a registered person is a customer, company rules may require that if the person is a customer, the person must be registered in the customer category, or the company may prohibit registration of anyone other than the customer on the mobile terminal 200 lent. As a result, all the people registered in the mobile terminal 200 may be determined to be customers.

When the first evaluation unit 130 analyzes the average number of interactions in a certain period for each customer as the amount of sales activity on the terminal 200 used by outstanding people and ordinary people, the frequency of communication with customers is analyzed. It becomes possible to analyze and make comparisons. Further, the first evaluation unit 130 may also analyze the time period during which communication is taking place. In this case, it becomes possible to analyze which time slots excellent people send messages and which time slots ordinary people send messages.

When the first evaluation unit 130 analyzes the number of times or percentage of use of trending words included in excellent communication information and general communication information over a certain period of time as a behavior pattern, it determines the extent to which trending words are used to communicate with customers. It will be possible to analyze and compare whether there are any

When the first evaluation unit 130 analyzes as a behavior pattern the percentage of replies with emoticons or stamps to the transmission of emoticons or stamps from customers included in the excellent communication information and general communication information, how much? It will be possible to analyze whether communication and interaction is carried out from a distance and use it as a target for comparison. Note that the analysis of pictograms and stamps may also be performed in consideration of the gender and age of the excellent person and the general person, and the customer's gender and age. In this case, for example, the analysis results may be displayed separately by the gender and age of the sales staff (with a range of ages such as 20s and 30s), or by the gender and age of the customer. The analysis results may also be displayed.

When the first evaluation unit 130 analyzes the average time to reply to a transmission from a customer in excellent communication information and general communication information, it analyzes the response speed and uses it as a comparison target. You will be able to do this.

When the first evaluation unit 130 analyzes the number of appearances or appearance rate of keywords indicating gratitude, joy, or requests from customers included in excellent communication information and general communication information, Or, it becomes possible to analyze whether a reply containing a keyword indicating a favor is received and use it as a target for comparison.

When the first evaluation unit 130 analyzes the number of transmissions or transmission ratio of keywords indicating compliance violation included in excellent communication information and general communication information over a certain period of time, the first evaluation unit 130 analyzes the number of times a term that constitutes compliance violation is used. You will be able to analyze and make comparisons.

When analyzing a target person, it becomes possible to analyze the position of the target person. For example, if you want to analyze yourself, you can input your own employee number from the input section 220 and compare and recognize your own results with the results for excellent people and the results for ordinary people for each item. They will be able to take appropriate actions such as changing their own sales style.

≪Method≫
Next, an example of an information processing method using the information processing device 100 will be described.

A target person is selected from the terminal 200 using an employee number or the like. At this time, the target period is selected by selecting the start time and end time. The start time and end time can be specified by, for example, the date or the date and time.

When the target person and the target period are input from the terminal 200, the control unit 50 reads out the evaluation results by the first evaluation unit 130 for the outstanding person, general person, and target person in the target period stored in the storage unit 60. It will be done. The evaluation results by the first evaluation section 130 are stored in the storage section 60, and the evaluation results stored in the storage section 60 may be read out by the control section 50.

Then, the first creation unit 110 creates visualization information about the evaluation results by the first evaluation unit 130 regarding the outstanding person and the general person read by the control unit 50 and the evaluation result by the first evaluation unit 130 regarding the target person. A document for explaining the visualized information is created by the second creation unit 120 and displayed on the display unit 210.

The information about excellent people, target people, and ordinary people stored in the storage unit 60 is created as follows, for example.

Information (for example, talk information) regarding interactions with customers on the terminals 200 held by excellent people and ordinary people is transmitted to the information processing device 100 in real time or at regular intervals (for example, once a day). These pieces of information are stored in the storage unit 60 together with the date and time in association with identification information for identifying each of the excellent person and the ordinary person. The number of customers registered in the terminals 200 held by excellent people and ordinary people is also transmitted to the information processing device 100 in real time or at a fixed period (for example, once a day).

The acquisition unit 10 acquires information regarding interactions with customers using the terminal 200 held by excellent persons and general persons stored in the storage unit 60 and the number of customers, together with identification information of the holder of the terminal 200. This acquisition may be performed in real time when the excellent person and the general person interact with the customer, or may be performed at a fixed period (for example, once a day).

When the acquisition unit 10 performs the acquisition, the first evaluation unit 130 analyzes information related to interactions with customers on the terminals 200 held by excellent individuals and ordinary individuals. The first evaluation unit 130 analyzes the amount of sales activities and behavior patterns described in the "structure" section above.

The amount of sales activity and behavior pattern analyzed by the first evaluation unit 130 are stored in the storage unit 60 together with the identification information of the owner of the terminal 200. Since the analysis results stored in the storage unit 60 are associated with identification information, by inputting the identification information from the input unit 220, the analysis results for each person in charge (excellent person and general person) can be obtained. can be read by the control unit 50.

As an example, the analysis results regarding excellent people are calculated as an average value. Similarly, analysis results regarding ordinary people are calculated as average values. Therefore, when a period is designated by the input unit 220, the average value of the analysis results regarding excellent people and the average value of the analysis results regarding ordinary people in the target period are read out by the control unit 50 for each item. This will be displayed on the display unit 210.

On the other hand, information regarding the target person can be obtained by inputting the identification information of the target person through the input unit 220, so that the control unit 50 reads out the analysis results regarding the target person for each period and displays them on the display unit 210. become.

The acquisition unit 10, search unit 20, comparison unit 40, first creation unit 110, second creation unit 120, first evaluation unit 130, second evaluation unit 140, control unit 50, etc. of this embodiment are integrated into one unit ( control unit) or a different unit. The functions of a plurality of "sections" may be integrated and realized by one unit (control unit). Further, the acquisition unit 10, the search unit 20, the comparison unit 40, the first creation unit 110, the second creation unit 120, the first evaluation unit 130, the second evaluation unit 140, the control unit 50, etc. may be realized by a circuit configuration. good.

The description of the embodiments described above and the disclosure of the drawings are merely examples for explaining the invention described in the claims, and the description of the embodiments and the disclosure of the drawings described above are only examples for explaining the invention described in the claims. This invention is not limited to this invention. Further, the description of the claims as originally filed is just an example, and the description of the claims may be changed as appropriate based on the description, drawings, etc.

10 Acquisition unit 60 Storage unit 70 Output unit 130 First evaluation unit 140 Second evaluation unit 200 Terminal

Claims (10)

An information processing device used for sales support,
an acquisition unit that acquires communication information regarding communication via a certain salesperson 's terminal;
Evaluating the characters included in the salesperson 's communication after performing natural language processing; Evaluating the number of times or the number of people with whom the salesperson communicated within a predetermined period; a first evaluation unit that evaluates the time required for a reply to a transmission from a customer in communication information or the time required for a reply from a customer to a transmission to a customer;
Equipped with
The acquisition department acquires excellent communication information via the terminal of a predetermined excellent person in charge of one or more sales who have received excellent evaluations,
The first evaluation department evaluates excellent communication information after performing natural language processing on the characters included in the communication, evaluates the number of times or number of people who communicated within a predetermined period, or even responds to customer transmissions. or the time it takes for a customer to respond to a message sent to a customer;
comprising an output unit that distinguishes and outputs both the evaluation result of the salesperson by the first evaluation unit and the evaluation result of excellent communication information by the first evaluation unit to the terminal of the salesperson ; Information processing device. The acquisition unit acquires general communication information via terminals of a plurality of ordinary people different from the excellent person,
The first evaluation department evaluates the general communication information after performing natural language processing on the characters included in the communication, evaluates the number of times or number of people who communicated within a predetermined period, or even responds to customer transmissions. or the time it takes for a customer to respond to a message sent to a customer;
The information processing device according to claim 1, wherein the output unit outputs the evaluation result of the general communication information by the first evaluation unit. The first evaluation department evaluates using the number of times or percentage of use of a trend word in a certain period of time, or the percentage of replies with pictograms or stamps to messages sent by customers with pictograms or stamps included in communication information. , The information processing device according to claim 1 or 2. According to any one of claims 1 to 3, the first evaluation unit performs the evaluation using the number of times or the percentage of appearance of a keyword in a certain period of time that indicates gratitude, joy, or request from the customer, which is included in the communication information. The information processing device described. The first evaluation section gives a high evaluation when the average time to reply to a transmission from a customer in communication information or the average time to a reply from a customer to a transmission to a customer is shorter than a threshold value, and gives a low evaluation when it is longer than the threshold value. The information processing device according to any one of claims 1 to 4, which is used for evaluation. The first evaluation unit determines keywords indicating compliance violations included in the communication information on the terminal,
The information processing device according to any one of claims 1 to 5, wherein if a keyword indicating compliance violation is included, a warning is sent to a terminal performing the communication. The information processing device according to any one of claims 1 to 6, further comprising a notification unit that recommends communication with the customer based on customer information. The information processing device according to any one of claims 1 to 7, further comprising a second evaluation unit that evaluates a disclosure level of personal information disclosed by a customer and gives a high evaluation to a customer with a high disclosure level. An information processing method used for sales support,
a step of acquiring, by an acquisition unit, communication information regarding interactions via a terminal of a certain salesperson ;
A first evaluation unit performs natural language processing on the characters included in the communication of the certain salesperson and evaluates the number of times or number of people with whom communication was conducted within a predetermined period of time, or Evaluating the time it takes to reply to a transmission from a customer or the time to a reply from a customer to a transmission to a customer in communication information of the certain salesperson ;
Equipped with
a step of acquiring excellent communication information via a terminal of a predetermined excellent person in charge of one or more sales who has received excellent evaluations by the acquisition department;
The first evaluation department evaluates excellent communication information after performing natural language processing on the characters included in the communication, evaluates the number of times or number of people who communicated within a specified period, or even responds to customer transmissions. or evaluating the time required for a customer to respond to a transmission to the customer;
A step of distinguishing and outputting both the evaluation result of the salesperson at the first evaluation section and the evaluation result of the excellent communication information by the first evaluation section to the terminal of the salesperson by the output section. An information processing method further comprising. A program installed on an information processing device,
The information processing device with the program installed is used for sales support,
A function to obtain communication information regarding interactions via a certain salesperson 's terminal,
Evaluating the characters included in the salesperson 's communication after performing natural language processing; Evaluating the number of times or the number of people with whom the salesperson communicated within a predetermined period; A function that evaluates the time it takes to reply to a transmission from a customer in communication information or the time to a reply from a customer to a transmission to a customer;
A function to acquire excellent communication information via a terminal of a predetermined excellent person in charge of one or more sales departments who have received excellent evaluations;
In excellent communication information, we evaluate the characters included in the communication after performing natural language processing, evaluate the number of times or number of people communicated within a specified period, or evaluate the time taken to reply to a transmission from a customer or the time it takes to reply to a customer. Ability to evaluate the time it takes for a customer to respond to a submission;
a function of distinguishing and outputting both the evaluation result of the salesperson by the first evaluation unit and the evaluation result of excellent communication information by the first evaluation unit to the terminal of the salesperson ;
A program that will equip you with the following.