JP7319478B1 - Information processing method, information processing system and program - Google Patents

Abstract

An object of the present invention is to efficiently detect a problem occurring in a company or a user. According to an information processing method, a computer generates a first log data group including a plurality of first log data and a second log data group including a plurality of second log data different in type from the first log data. obtaining a support content data group including a plurality of support content data, and obtaining first keyword information corresponding to one support content data in the support content data group from the first log data group and the second log data group. extracting first log data and second log data including including extracting problematic keyword information from a plurality of pieces of first keyword information included in . [Selection drawing] Fig. 1

Description

The present invention relates to an information processing method, an information processing system and a program.

Communication may be arranged between a company and a customer in order to solve a problem that the company or customer has encountered. In order to improve operational efficiency, support content (also referred to as support content data) may be provided so that customers can solve problems without directly communicating with the company. An example of support content data is FAQ (Frequently Asked Questions).

Also, for the purpose of generating the above-mentioned FAQ and improving services, information on communication between the company and the customer may be recorded. For example, call information (call log data) between a user and an operator is one example. Patent Literature 1 discloses analyzing call log data.

International Publication No. 2014-141452

On the other hand, in addition to call log data, there is communication information between users (social log data) as information from users regarding problems. However, since call log data and social log data are different types of information, it is necessary for the person in charge to manually acquire and analyze the relevant information, which is time-consuming and burdensome.

One of the objects of the present invention is to efficiently detect problems occurring in companies and users.

According to one embodiment of the present invention, a computer comprises a first log data group including a plurality of first log data and a second log data group including a plurality of second log data different in type from the first log data. and acquiring a support content data group including a plurality of support content data, and obtaining a first keyword corresponding to one support content data in the support content data group from the first log data group and the second log data group. Extracting first log data and second log data containing information to generate integrated data, and performing the integration based on a numerical relationship between support content data and at least one of the first log data and the second log data An information processing method is provided that includes extracting problematic keyword information from among a plurality of pieces of first keyword information included in data.

In the above information processing method, a ratio of support content data corresponding to the problem keyword information in the support content data group, a ratio of first log data corresponding to the problem keyword information in the first log data group, and Problem information for the problem keyword information may be generated based on a relationship with a ratio of second log data corresponding to the problem keyword information in the second log data group.

In the above information processing method, a ratio of support content data corresponding to the problem keyword information in the support content data group, a ratio of first log data corresponding to the problem keyword information in the first log data group, and First priority information for the problematic keyword information may be generated based on a relationship with a ratio of second log data corresponding to the problematic keyword information in the second log data group.

In the above information processing method, a ratio of support content data corresponding to the problem keyword information in the support content data group is lower than a first threshold, and a first log corresponding to the problem keyword information in the first log data group. When the ratio of data is higher than the second threshold, first priority information higher than other problem keyword information may be generated.

In the above information processing method, second priority information for the problem keyword information is generated based on the number of data in the first log data for the problem keyword information, and the first priority information and the second priority information are: Based on this, third priority information for the problem keyword information may be generated.

In the above information processing method, among the plurality of pieces of first keyword information, first keyword information that satisfies a predetermined condition in terms of a ratio of the first log data to the first log data group may be excluded.

In the above information processing method, the first keyword information may be updated based on trend information acquired at predetermined intervals.

In the above information processing method, the first priority information may be generated based on at least one of trend information and user attribute information acquired at predetermined intervals.

In the above information processing method, first log data corresponding to second keyword information having a low degree of association with the first keyword information may be excluded from the first log data group.

In the above information processing method, first log data corresponding to second keyword information having a low degree of association with the first keyword information may be excluded from the first log data group.

In the above information processing method, the first log data includes one of call log data, social log data, and chat log data, and the second log data includes the other of call log data, social log data, and chat log data. may include one of

According to one embodiment of the present invention, a first log data group including a plurality of first log data; a second log data group including a plurality of second log data different in type from the first log data; obtaining a support content data group including support content data of the above, and including first keyword information corresponding to one support content data in the support content data group from the first log data group and the second log data group Extracting the first log data and the second log data to generate integrated data, and based on a numerical relationship between the support content data and at least one of the first log data and the second log data, included in the integrated data An information processing system is provided that includes a control unit that extracts problematic keyword information from among a plurality of pieces of first keyword information.

In the above information processing method, the control unit controls the proportion of support content data corresponding to the problem keyword information in the support content data group, and a first log corresponding to the problem keyword information in the first log data group. The problem information for the problem keyword information may be generated based on the ratio of the data and the ratio of the second log data corresponding to the problem keyword information in the second log data group.

In the above information processing method, the control unit controls the proportion of support content data corresponding to the problem keyword information in the support content data group, and a first log corresponding to the problem keyword information in the first log data group. The first priority information for the problematic keyword information may be generated based on the ratio of the data and the ratio of the second log data corresponding to the problematic keyword information in the second log data group.

In the above information processing method, the control unit is configured such that a ratio of support content data corresponding to the problem keyword information in the support content data group is lower than a first threshold, and the problem keyword information in the first log data group When the ratio of corresponding first log data is higher than the second threshold, first priority information higher than other problem keyword information may be generated.

In the above information processing method, the control unit generates second priority information for the problem keyword information based on the number of data in the first log data for the problem keyword information, and generates the first priority information and the first log data for the problem keyword information. Third priority information for the problem keyword information may be generated based on the second priority information.

In the above information processing method, the control unit may exclude, from among the plurality of pieces of first keyword information, first keyword information satisfying a predetermined condition in a ratio of the first log data to the first log data group. .

In the information processing method described above, the control unit may update the first keyword information based on trend information acquired at predetermined intervals.

In the information processing method described above, the control unit may generate the first priority information based on at least one of trend information and user attribute information acquired at predetermined intervals.

In the information processing method described above, the control unit may exclude, from the first log data group, first log data corresponding to second keyword information having a low degree of association with the first keyword information.

In the information processing method described above, the control unit may update the second keyword information based on trend information acquired at predetermined intervals.

In the information processing method described above, the control unit may configure the first log data to include one of call log data, social log data, and chat log data, and the second log data to include call log data, social log data, and chat log data. It may also include other one of the log data.

According to one embodiment of the present invention, a computer stores a first log data group including a plurality of first log data and a second log data group including a plurality of second log data of a different type from the first log data. and acquiring a support content data group including a plurality of support content data, and obtaining a first keyword corresponding to one support content data in the support content data group from the first log data group and the second log data group. Extracting first log data and second log data containing information to generate integrated data, and performing the integration based on a numerical relationship between support content data and at least one of the first log data and the second log data A program is provided for extracting problematic keyword information from a plurality of pieces of first keyword information included in data.

ADVANTAGE OF THE INVENTION According to this invention, the problem which arises in a company or a user can be detected efficiently.

It is a figure explaining the information integration processing system in one Embodiment of this invention. It is a figure explaining the hardware constitutions of each apparatus which comprises the information integration processing system in one Embodiment of this invention. 4 is a diagram illustrating the software configuration of the control unit of the information integration server in one embodiment of the present invention; FIG. It is a figure which shows the exclusion keyword table in one Embodiment of this invention. It is a figure which shows the integrated keyword table in one Embodiment of this invention. FIG. 4 is a diagram showing an association table of keywords and log data in one embodiment of the present invention; It is a figure which shows the integrated data table in one Embodiment of this invention. It is a figure which shows the ratio data table in one Embodiment of this invention. It is a figure which shows the correspondence priority table in one Embodiment of this invention. It is a figure which shows the problem-solving data table in one Embodiment of this invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. 4 is an integrated data set showing the number and ratio of call log data for each integrated keyword in one embodiment of the present invention. It is a flow figure showing information integration processing of the information integration processing system in one embodiment of the present invention. It is an integrated data set and a priority data table in one embodiment of the present invention.

The embodiments shown below are examples of embodiments of the present invention, and the present invention is not limited to these embodiments. In the drawings referred to in this embodiment, the same parts or parts having similar functions are given the same reference numerals or similar reference numerals, and repeated description thereof may be omitted.

<First Embodiment>
An information integration processing system according to a first embodiment of the present invention will be described in detail with reference to the drawings.

(1-1. Overall configuration of information integration processing system)
FIG. 1 is a diagram for explaining an information integration processing system 1 according to the first embodiment of the present invention. This information integration processing system 1 includes an information integration server 10 (information providing device), a call center CC, a web server 30, a user terminal 40, and a company side terminal 60. FIG.

A database 15 is connected to the information integration server 10 . The call center CC includes a call center server 20, database 25, and operator terminal 50. FIG. Although one operator terminal 50 is provided in FIG. 1, two or more operator terminals 50 may be provided.

In FIG. 1, two networks 91, 93 are shown. A network 91 is a communication network such as the Internet that uses a packet switching method. The network 93 is a public switched telephone network (PSTN) using a line switching system such as a telephone line. The user terminal 40 is a terminal such as a smart phone that can at least make calls and browse web pages, and is connectable to networks 91 and 93 . A network 91 is used for communication using a web browser, and a network 93 is used for communication by telephone. The information integration server 10 , call center server 20 , web server 30 and company terminal 60 are connected to a network 91 .

The call center CC may be provided with a private telephone network and a LAN (Local Area Network) (not shown). In the call center CC, the telephone line used by the operator terminal 50 may be controlled by a PBX (Private Branch Exchange) (not shown).

According to the information integrated processing system 1, a user browses a web page by accessing the web server 30 using the user terminal 40, and initiates a call command embedded in the web page to make a call to the call center CC. You can call and talk to the operator. Further, according to the information integration processing system 1, the user can input a message to the SNS site by accessing the web server 30. FIG. Call records (call log data) between the user and the operator and messages (social log data) input to the SNS are integrated by the information integration server 10 together with support content data for each keyword. Supporting content data means information that helps the user. In this embodiment, FAQ data is used as support content data. Further, the information integration server 10 generates problem information based on the integrated data, and also generates solution information corresponding to the priority of response and problem information, and the generated information is displayed on the company side terminal 60 . The configuration of an information integration processing system for realizing such processing will be described below.

(1-2. Hardware configuration)
FIG. 2 is a diagram for explaining the hardware configuration of each device that constitutes the information integration processing system 1. As shown in FIG.

(1-2-1. Information Integration Server 10)
Information integration server 10 includes control unit 11 , storage unit 12 and communication unit 13 . The control unit 11 is an example of a computer including arithmetic processing circuits such as a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), and an FPGA (Field Programmable Gate Array). The control unit 11 executes the programs stored in the storage unit 12 to implement various functions in the information integration server 10 . As a result, processing (information integration processing) by the information integration server 10 in the information integration processing system 1 is executed.

The storage unit 12 includes a storage device and stores control programs. The storage unit 12 also stores various data used when the control program is executed. Note that this program may be provided in a state recorded in a computer-readable recording medium such as a magnetic recording medium, an optical recording medium, a magneto-optical recording medium, or a semiconductor memory. In this case, the information integration server 10 may include an interface for connecting recording media. Here, the storage medium may be specified as a medium separate from the storage unit 12 included in the information integration server 10 or may be a medium used for the storage unit 12 .

The communication unit 13 includes a communication module, is connected to the network 91 under the control of the control unit 11 , and transmits and receives information to and from other devices connected to the network 91 . In this example, the communication unit 13 is also connected to the database 15 to transmit and receive information. Information registered in the database 15 will be described later. Note that the communication unit 13 may be connected to the database 15 via the network 91 . Information registered in the database 15 may be stored in the storage unit 12 . In this case, database 15 may not exist.

(1-2-2. Call center server 20)
Call center server 20 includes control unit 21 , storage unit 22 and communication unit 23 . The control unit 21 has basically the same configuration as the control unit 11 described above, and causes the call center server 20 to implement various functions. Thereby, the processing by the call center server 20 in the information integrated processing system 1 is executed.

The storage unit 22 has basically the same configuration as the storage unit 12, and the contents of the instructions of the stored programs are different, so the description thereof will be omitted. The communication unit 23 has basically the same configuration as that of the communication unit 13, and is connected to the database 25 only. The database 25 may store user information and call log data. Note that the communication unit 23 may be connected to the database 25 via the network 91 . Information registered in the database 25 may be stored in the storage unit 22 . In this case, database 25 may not exist. Various types of acquired information may be stored in a separately provided database.

(1-2-3. Web server 30)
Web server 30 includes control unit 31 , storage unit 32 and communication unit 33 . The control unit 31 has basically the same configuration as the control unit 11 described above, and causes the Web server 30 to implement various functions. As a result, processing by the web server 30 in the information integrated processing system 1 is executed.

The storage unit 32 has basically the same configuration as the storage unit 12, and the contents of the instructions of the stored programs are different, so the description thereof will be omitted. Since the communication section 33 has basically the same configuration as the communication section 13, the description thereof will be omitted.

(1-2-4. User terminal 40)
User terminal 40 includes control unit 41 , storage unit 42 , communication unit 43 , display unit 44 , operation unit 45 and call unit 46 . The control unit 41 has basically the same configuration as the control unit 11 described above, and causes the user terminal 40 to implement various functions. Processing in the user terminal 40 in the information integrated processing system 1 is executed.

The storage unit 42 has basically the same configuration as the storage unit 12, and the contents of the instructions of the stored programs are different, so the description thereof is omitted. The communication unit 43 has basically the same configuration as that of the communication unit 13, and the connectable networks are different, so the description thereof will be omitted.

The display unit 44 includes a display device whose display contents are controlled under the control of the control unit 41 . The operation unit 45 includes a touch sensor in this example, and outputs information corresponding to the position operated by the user to the control unit 41 . This touch sensor is provided on the display area of the display unit 44 . That is, the display unit 44 and the operation unit 45 constitute a touch panel. The calling unit 46 includes functions necessary for a telephone call, and includes, for example, a speaker that outputs signals supplied from the control unit 41 as sounds, and a microphone that outputs input sound signals to the control unit 41 .

(1-2-5. Operator terminal 50)
Operator terminal 50 includes control unit 51 , storage unit 52 , communication unit 53 , display unit 54 , operation unit 55 and call unit 56 . The control unit 51 has basically the same configuration as the control unit 11 described above, and causes the operator terminal 50 to implement various functions. Thereby, the processing in the operator terminal 50 in the information integrated processing system 1 is executed.

The storage unit 52 has basically the same configuration as the storage unit 12, and the contents of the stored program instructions are different, so the description thereof is omitted. The communication unit 53 has basically the same configuration as that of the communication unit 13, and the connectable networks are different, so the description thereof is omitted.

The display unit 54 includes a display device whose display content is controlled under the control of the control unit 51 . The operation unit 55 includes a keyboard in this example, and outputs information corresponding to the button (position) pressed by the operator's finger to the control unit 51 . The call unit 56 is configured by a headset in this example. Call unit 56 includes a speaker that outputs a signal supplied from control unit 51 as sound, and a microphone that outputs an input sound signal to control unit 51 .

(1-2-6. Company side terminal 60)
The enterprise terminal 60 includes a control section 61 , a storage section 62 , a communication section 63 , a display section 64 and an operation section 65 . The control unit 61 has basically the same configuration as the control unit 11 described above, and implements various functions in the company-side terminal 60 . The processing in the company side terminal 60 in the information integrated processing system 1 is executed.

The storage unit 62 has basically the same configuration as the storage unit 12, and the contents of the instructions of the stored programs are different, so the description thereof will be omitted. The communication unit 63 has basically the same configuration as that of the communication unit 13, and the connectable networks are different, so the description thereof is omitted. The display unit 64 has basically the same configuration as the display unit 44, and the contents displayed are different, so the description thereof will be omitted. The operation unit 65 has basically the same configuration as the operation unit 45, and the operation contents are different, so the description thereof is omitted.

(1-3. Software configuration of information integration server)
FIG. 3 is a diagram showing the software configuration of the control unit 11 of the information integration server 10. As shown in FIG. The control unit 11 includes an acquisition unit 11a, a data organization unit 11b, a data integration unit 11c, an analysis unit 11d, and a generation unit 11e.

The acquisition unit 11a has a function of acquiring various types of information from each device. In this example, the FAQ data are stored in the FAQ database 15a. Call log data (first log data) is stored in the call log database 15b. Social log data (second log data) is stored in the social log database 15c. In this embodiment, a plurality of FAQ data stored in the FAQ database 15a is also called an FAQ data group. A plurality of call log data stored in the call log database 15b is also referred to as a call log data group (first log data group). A plurality of social log data stored in the social log database 15c is also called a social log data group (second log data group).

The data organizing unit 11b has a function of excluding (organizing) data that meets a predetermined condition from the acquired log data. In this example, the data organizing unit 11b excludes call log data or social log data including an exclusion keyword (second keyword) from the call log data group and the social log data group.

The data integration unit 11c has a function of integrating FAQ data, call log data, and social log data. In this example, the data integration unit 11c forms an integrated data table by associating FAQ data, call log data, and social log data with corresponding keywords among multiple integrated keywords (first keywords). Specifically, the data integration unit 11c extracts from the call log data group and the social log data group at least one of first log data and second log data containing an integrated keyword corresponding to one FAQ data in the FAQ data group. to generate integrated data.

The analysis unit 11d analyzes the integrated data table based on predetermined conditions. In this example, the ratio of FAQ data, call log data, and social log data for each integrated keyword is calculated, and the data ratio is used as a threshold for comparison and analysis. Further, the analysis unit 11d extracts problematic keyword information based on the numerical relationship between the call log data or the social log data and the FAQ data from the integrated data.

The generation unit 11e has a function of generating, from the integrated data table, problem information indicating specific problem content for problem keyword information, response priority information indicating priority corresponding to the problem, and solution information for the problem.

(1-4. Various tables)
Next, various data tables used in the information integration processing system 1 will be described.

(1-4-1. Exclusion keyword table)
FIG. 4 is an exclusion keyword table 100. As shown in FIG. Excluded keywords may be set in advance. As shown in FIG. 4, exclusion keywords may include information such as news, BOT accounts, continuous posts, retweets, tweets with the same first 20 characters, and promotional tweets, for example. Excluded keywords may be periodically changed based on trend information obtained at predetermined intervals. The information integration server 10 determines whether or not each call log data contains an exclusion keyword. If call log data includes an exclusion keyword, the call log data is excluded. Note that the excluded call log data may be stored in the database 15 as it is. By executing this processing, it is possible to extract the pure voice of the user, so that the quality of the data can be improved. As a result, it is possible to accurately detect problems from the acquired data.

(1-4-2. Integrated keyword table)
FIG. 5 is an integrated keyword table 200. As shown in FIG. Integrated keyword table 200 may be set in advance. For example, integrated keywords (also referred to as “first keywords” and “intermediate keywords”) may be set based on FAQ categories. Also, the integrated keyword table 200 may be updated (regularly) at predetermined intervals.

(1-4-3. Association data table)
FIG. 6 is an association data table 300 that associates an integrated keyword with each piece of log data. The association data table 300 includes integrated keywords 310 and log data 320 . In the association data table 300, the integrated keywords 310 are arranged horizontally. The log data 320 are arranged vertically.

(1-4-4. Integrated data table)
FIG. 7 is the integrated data table 400. As shown in FIG. The integrated data table 400 includes summary data 410 of data integrated information and extracted data 420 extracted from each acquired data. Summary data 410 includes integrated ID 411 , integrated keyword 413 , and data source 415 . In the extracted data 420, FAQ ID 421a and FAQ title information 421b are extracted from the FAQ data and arranged in the integrated data table. From the call log data, a data creation date 423b and a subject 423c of the call log data are extracted together with the call log ID 423a and arranged in the integrated data table. From the social log data, together with the call log ID 425a, the date and time of data generation 425b and message information 425c are extracted and arranged in the integrated data table.

(1-4-5. Ratio data table)
FIG. 8 is a ratio data table 500. As shown in FIG. The ratio data table 500 includes an integrated keyword 510, the number of FAQs 520a and the ratio 520b corresponding to each of the integrated keywords, the number of call log data 530a and the ratio 530b corresponding to each of the integrated keywords, and the social data corresponding to each of the integrated keywords. Includes log data count 540a and percentage 540b.

(1-4-5. Correspondence priority table)
FIG. 9 is the corresponding priority data table 600. As shown in FIG. Correspondence priority data table 600 includes keyword information 605 , pattern ID 610 , FAQ data ratio 620 for keywords, call log data ratio 630 , social log data ratio 640 , and correspondence priority information 650 .

(1-4-6. Problem solving data table)
FIG. 10 is a problem solving data table 700. FIG. The problem solving data table 700 includes keyword information 705 , pattern ID 710 , FAQ data ratio 720 for keywords, call log data ratio 730 , social log data ratio 740 , problem information 750 , response priority 760 , and solution information 770 .

(1-5. Information integration processing)
Next, information integration processing implemented in the information integration processing system according to the first embodiment of the present invention will be described. 11 to 16 are flowcharts showing information integration processing of the information integration processing system 1 according to this embodiment.

(1-5-1. Acquisition of call log data)
First, in the flow of FIG. 11, the user inputs login information to the user terminal 40 (step S101). The login information is transmitted to the web server 30, and the web page request and response are repeated (step S103), and the web page is displayed on the display unit 44. FIG. When the Web page of the inquiry screen is requested from the user terminal 40 to the Web server 30, the Web server 30 transmits data of the inquiry screen to the user terminal 40 (step S103). When the user terminal 40 receives this data, an inquiry screen is displayed on the display section 44 .

At this time, when the user presses a call request button on the inquiry screen (makes a call request), call request information is transmitted from the Web server 30 to the call center server 20 (step S105), and the call center server 20 sends a call instruction. is executed (step S107). By activating this program, the operator terminal 50 to talk with the user terminal 40 is determined via the PBX (step S109), and a call is made between the user terminal 40 and the operator terminal 50 (step S111). At this time, the call center server 20 generates a call record (call log data) (step S113). The call log data is stored in the database 25 and transmitted to the information integration server 10 (step S115). The information integration server 10 acquires the call log data (step S117) and stores it in the database 15. FIG.

(1-5-2. Acquisition of social log data)
In the flow of FIG. 12, the user inputs login information to the user terminal 40 (step S201). The login information is transmitted to the web server 30 (step S203), and the SNS (social network site) web page is displayed on the display unit 44. FIG. At this time, SNS application software (app) may be used. When the user inputs a message via the operation unit 45 of the user terminal 40, the web server 30 generates social log data corresponding to the message (step S205). The web server 30 stores the social log data in the storage unit 32 and also transmits it to the information integration server 10 (step S207). The information integration server 10 acquires the transmitted social log data (step S209) and stores it in the database 15. FIG.

In this embodiment, the processes shown in FIGS. 11 and 12 are sequentially repeated, so call log data and social log data are sequentially acquired. As a result, a call log data group including a plurality of call log data and a social log data group including a plurality of social log data are generated. At this time, the acquisition timing of each log data is not particularly limited.

In this embodiment, the FAQ data may be stored in the database 15 in advance, or may be acquired at this timing.

(1-5-3. Arrangement of acquired data)
As shown in FIG. 13, the information integration server 10 excludes call log data corresponding to exclusion keywords included in the exclusion keyword table 100 shown in FIG. 4 from the call log data group (step S301). In this example, the information integration server 10 excludes call log data corresponding to the exclusion keyword from the call log data group including a plurality of call log data.

Similarly, the information integration server 10 excludes social data corresponding to exclusion keywords corresponding to exclusion keywords included in the exclusion keyword table 100 shown in FIG. 4 from the social data group (step S303).

As an exclusion keyword, a keyword having low relevance to the integrated keyword may be set. As a result, unnecessary log data can be excluded, and problems can be detected more efficiently.

(1-5-4. Integration of data)
As shown in FIG. 14, the information integration server 10 classifies FAQ data by integrated keyword (step S401).

Next, the information integration server 10 classifies each log data of the call log data and the social log data for each integrated keyword corresponding to the FAQ data (steps S403, S405). FIG. 15 shows a method of classifying call log data and social log data.

As shown in FIG. 15, the information integration server 10 collates each log data with each integrated keyword (step S4031). At this time, if an integrated keyword is included in each data (S4033; Yes), as shown in FIG. 6, a flag ("1" in this example) is set corresponding to the integrated keyword (step S4035). . If no integrated keyword is included in each log data (S4033; No), no flag is set (step S4037) (or a flag is set corresponding to the keyword "other").

As a result of repeating the above process, the information integration server 10 extracts and integrates FAQ data, flagged call log data, and social log data in association with keywords, as shown in FIG. 7 (step S407). Thereby, the integrated data table 400 is generated. Integrated IDs do not have to be set for each data group for FAQ data, call log data, and social log data. For example, the integrated ID may be set according to the generation date or generation time of the log data. By referring to FIG. 7, the user can collectively understand what keywords each data is associated with using one data table. By referring to this integrated data table, the user can easily grasp what kind of problem has occurred, and can efficiently solve the problem.

(1-5-5. Analysis of data)
In the flow of FIG. 16, the information integration server 10 calculates the ratio of each integrated keyword in each data group (step S501). In this example, as shown in FIG. 8, when the keyword is "login", the total number of FAQs is 95, whereas the number of FAQs with the keyword "login" is 2, and the ratio of FAQs is 2. .1%. Similarly, the total number of call log data is 802, and the number of call log data for the keyword "login" is 154. Therefore, the ratio of call log data for the keyword "login" is 19.2%. . Similarly, the total number of social log data is 118, and the number of social log data for the keyword "login" is 3. Therefore, the ratio of social log data for the keyword "login" is 2.5%. .

Returning to FIG. 16, description will be made. The information integration server 10 performs analysis processing using the calculated ratio of each data to each keyword (step S503). In this example, as shown in FIG. 9, each data ratio for each keyword is analyzed (determined) as data ratio information “high (high)” and “low (low)”. At this time, a threshold value may be provided to analyze "high" and "low" of each data ratio for each keyword. The threshold may be the same or different in the data ratio of each data. For example, the threshold for the FAQ rate may be "9.5%", the threshold for the call log data rate may be "8.2%", and the threshold for the social log data rate may be "11.8%". At this time, as shown in FIG. 9, in this embodiment, eight patterns A to H are set as analysis patterns using the ratio of each data to one keyword.

Returning to FIG. 16, description will be made. The information integration server 10 extracts the problematic integrated keyword (problem keyword information) based on the above analysis result, that is, the numerical relationship between the FAQ data, the call log data, and the social log data. At this time, the information integration server 10 also generates problem information corresponding to the problem integrated keyword (step S505). In this example, as shown in FIG. 10, at this time, problem information may be generated based on the relationship between the FAQ data rate, the call log data rate, and the social log data rate. In the case of pattern A (FAQ: high, call log data: high, social log data: high), "FAQ site configuration (delivery rate) and individual FAQ page content (solution rate)" Generate problem information that there is a problem. Similarly, in the case of pattern E (FAQ: low, call log data: high, social log data: high), problem information indicating that there is a problem with the "individual FAQ quantity" is generated. Note that the problem information may be generated based on the relationship between the FAQ data rate and the call log data rate or social log data rate according to conditions.

Returning to FIG. 16, description will be made. Further, the information integration server 10 generates response priority information 650 for the problem (problem keyword information) corresponding to the analysis pattern (condition) (step S507). At this time, the correspondence priority information 650 may be generated based on the relationship between the FAQ data ratio and the call log data ratio or social log data ratio. For example, as shown in FIG. 9, when the FAQ data ratio is high and the call log data ratio or social log data ratio is low, the priority may be set to be high. Specifically, when the FAQ data rate is "low" and the call log data rate and social log data rate are "high", the highest correspondence priority information (priority "8" in this example) is generated. Also, when the FAQ data ratio is high and the call log data ratio or social log data ratio is low, the priority may be set to be low. Specifically, when the FAQ data is "high" and the call log data ratio and social log data ratio are "low", the lowest response priority information (priority "8" in this example) is generated (set). .

Returning to FIG. 16, description will be made. The information integration server 10 generates solution information for each problem (step S509). In this example, as shown in the problem solution data table 700 of FIG. 10, the information integration server 10 generates solution information 770 corresponding to the problem. For example, for the problem of "individual FAQ quantity", information on FAQ/knowledge creation (rewriting) support service, UI (User Interface)/UX (User Experience) (quantitative and qualitative) research, and LITIFA analysis service is provided. Similarly, for the problem of FAQ site configuration (delivery rate), information on UI / UX (quantitative and qualitative) research, addition and improvement of channels (chat, chat bot), social listening, and individual FAQ page content (solution) We provide information on FAQ assessment services, social listening, and FAQ/knowledge production support services in response to the problem of the rate of production.

The information integration server 10 transmits the generated problem information, response priority information, and solution information to the terminal of the person in charge of the company (company-side terminal 60) (step S511). The problem solving information is displayed on the display unit 64 of the company terminal 60 (step S513). This allows company personnel to easily know how to modify the FAQ.

By using this embodiment, various types of information can be integrated and the information can be unified, so that problems can be easily grasped. Further, by using this embodiment, it is possible to generate information that could not be obtained from one type of information, and to efficiently detect problems occurring in companies and users. Furthermore, by using this embodiment, a priority is set for each data, and a solution according to the priority is provided, so that the problem can be solved quickly.

<Second embodiment>
In this embodiment, an example will be described in which integrated keyword information determined to be abnormal is excluded before priority information is generated from integrated data.

FIG. 17 is a flow regarding exclusion of integrated keywords. FIG. 18 is an integrated data set 500A showing the number 530Aa and the ratio 530Ab of call log data for each integrated keyword 510A.

First, the information integration server 10 calculates the average value and standard deviation for the number of call log data for each integrated keyword (step S601). Subsequently, the information integration server 10 calculates an upper limit value and a lower limit value for deviation from the calculated average value and standard deviation (step S603).

Specifically, as shown in FIG. 18, an average value (Ave.) of 422.5 and a standard deviation (σ) of 1138.2 are calculated from the number of call log data for each integrated keyword. At this time, the deviation upper limit value is set to the average value+3×σ. As a result, 3837.0 is calculated as the deviation upper limit in this embodiment. Similarly, the outlier lower limit (average value -3σ) is calculated as -2991.9.

Next, the information integration server 10 determines whether the number of data items for each integrated keyword in the integrated data set 500A exceeds the outlier upper limit or falls below the outlier lower limit (step S605). If the number of data is equal to or more than the lower limit value of outliers and equal to or less than the upper limit value of outliers (step S605; No), the process ends without performing the exclusion process. On the other hand, if the number of data exceeds the outlier upper limit or falls below the outlier lower limit (step S605; Yes), the corresponding integrated keyword is excluded (step S607). In this example, the integrated keyword “withdrawal” is 3846, which exceeds the upper limit of deviation (3837.0). As a result, as shown in FIG. 18, "withdrawal" is removed from the integrated keyword list.

By using this embodiment, abnormal data can be excluded, so priority information can be generated with higher accuracy, and problematic information can be detected (extracted).

Although the number of call log data items has been described in the present embodiment, the present invention is not limited to this. Similar processing may be performed on social log data and FAQ data.

<Third Embodiment>
In the first embodiment of the present invention, an example of generating one piece of priority information was shown, but in this embodiment, an example of setting corresponding priority from a plurality of pieces of priority information will be described.

FIG. 19 is a flowchart for setting correspondence priority. FIG. 20 shows an integrated data set 500B and a priority data table showing FAQ data count 520Ba and ratio 520Bb, call log data count 530Ba and ratio 530Bb, social log data count 540Ba and ratio 540Bb for each integrated keyword 510B. 800. The priority data table 800 includes an integrated keyword 810, first priority 820, second priority 830, and corresponding priority 840 (third priority).

First, as shown in FIG. 9, the first priority 820 is set (generated) for each integrated keyword from the correspondence priority data table 600 (step S701). Next, the information integration server 10 determines whether or not there is an integrated keyword with the same first priority (step S703). When there is no integrated keyword having the same first priority (step S703; No), the corresponding priority 840 (third priority) is directly set (generated) (step S711).

On the other hand, when there are a plurality of integrated keywords having the same first priority (step S703; Yes), the integrated keywords having the same first priority are prioritized based on the ratio of call log data. 2 Priority 830 is generated (step S705). Furthermore, if the ratio of call log data is the same, the second priority 830 is set (generated) based on the ratio of social log data to the integrated keyword having the same first priority (step S707). .

The information integration server 10 sets (generates) the correspondence priority 840 (third priority) based on the first priority 820 and the second priority 830 (step S709).

By using this embodiment, it is possible to set the response priority with higher accuracy, and to more reliably extract the problem to be addressed.

In this embodiment, when the first priority is the same and the data ratio of call log data and social log is the same, additional priority is added based on the ratio of support content data (FAQ data). may be generated.

<Modification>
The present disclosure is not limited to the embodiments described above, and includes various other modifications. For example, the above-described embodiments have been described in detail in order to explain the present disclosure in an easy-to-understand manner, and are not necessarily limited to those having all the described configurations. Also, part of the configuration of one embodiment may be replaced with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Moreover, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration. Some modifications will be described below. It should be noted that the modified example of the first embodiment can also be applied as a modified example of other embodiments.

In the above-described embodiment, the functions of each server that constitutes the information integration processing system may be implemented in cooperation with a plurality of servers. Also, the functions of each server may be integrated into one server. For example, the web server 30 and the information integration server 10 may be realized by the same server. The Web server 30 and the call center server 20 are preferably separate servers for security purposes due to differences in the types of data handled. may be one server.

Note that in one embodiment of the present invention, the information integration server 10 may not necessarily be provided. In this case, similar processing can be executed in the control unit 61 of the company terminal 60 . Also, in one embodiment of the present invention, application software such as Microsoft's Excel or Access may be used.

In the embodiments described above, the evaluation result information and the problem solving information may be transmitted to the user terminal at the same timing.

In the first embodiment of the present invention, an example of integrating call log data and social log data together with FAQ data was shown, but the present invention is not limited to this. For example, in addition to call log data and social log data, chat data may be acquired and integrated with FAQ data (support content data). Further, either call log data or social log data and chat data may be obtained and integrated with the FAQ data. Furthermore, other different types of data and FAQ data may be integrated.

Further, in the first embodiment of the present invention, FAQ is used as support content data, but the present invention is not limited to this. For example, it may be information showing how to use the service, an online manual, an explanatory video, or the like.

Also, the integrated keyword may be updated based on trend information acquired at predetermined intervals. For example, the integrated keyword may be updated based on ranking information calculated from the number of data in the call log data group or the social log data group (second log data group). This allows you to set the best keywords according to current trends. Therefore, it can be useful for generating necessary FAQs.

Also, in the first embodiment of the present invention, an example in which integrated keywords are set based on FAQ categories has been described, but the present invention is not limited to this. Integrated keywords may be set based on user attributes (industry information, customer information). As a result, it is possible to generate integrated data according to the attributes of the user, and to extract and solve problems according to the attributes of the user.

Also, the integrated keyword may be appropriately updated/generated by machine learning. This makes it possible to always optimize integrated keywords, making it easier to extract problems to be solved.

Also, the integrated keyword may be generated from the degree of similarity between support content information, information such as social log data (also referred to as public information), and call log data.

Also, in the first embodiment of the present invention, an example was shown in which each data ratio is analyzed as "high" and "low", but the present invention is not limited to this. For example, it may be evaluated and analyzed as "high", "medium", and "low" according to the ratio, or may be evaluated and analyzed in five stages from "1" to "5" according to the ratio.

Also, the correspondence priority may be changed according to the importance of the keyword. The degree of importance of a keyword may be set in advance, or may be set according to the number of call log data or the number of social log data. For example, if the priority is "3" and the keyword's importance is "2" from "1" (high importance) to "5" (low importance), the priority is raised by one. You can change it to "2". As a result, it is possible to respond according to the importance of the keyword.

In the first embodiment of the present invention, the order and timing of generating priority information and problem information may be changed as appropriate.

Also, in the first embodiment of the present invention, an example of calculating the FAQ data ratio, the call log data ratio, and the social log data ratio was shown, but the present invention is not limited to this. It may be the number of FAQ data, the number of call log data, and the number of social log data corresponding to one integrated keyword, or if the numerical relationship between FAQ data and call log data or social log can be expressed, Other values may be used.

In the first embodiment of the present invention, an example of using a priority data set to generate priority information was shown, but the present invention is not limited to this. For example, weighting may be set so as to give higher priority to specific keyword information in accordance with user attributes. For example, when the user attribute is "new customer", a specific integrated keyword may be given a higher priority than other integrated keywords.

DESCRIPTION OF SYMBOLS 1... Information integration processing system, 10... Information integration server, 11... Control part, 11a... Acquisition part, 11b... Data arrangement part, 11c... Data integration part, 11d... Analysis unit 11e Generation unit 12 Storage unit 13 Communication unit 15 Database 15a Database 15b Call log database 15c Social Log database 20 Call center server 21 Control unit 22 Storage unit 23 Communication unit 25 Database 30 Server 31 Control unit 32... Storage unit, 33... Communication unit, 40... User terminal, 41... Control unit, 42... Storage unit, 43... Communication unit, 44... Display unit, 45 Operation unit 46 Communication unit 50 Operator terminal 51 Control unit 52 Storage unit 53 Communication unit 54 Display unit 55 Operation unit 56 Communication unit 60 Company terminal 61 Control unit 62 Storage unit 63 Communication unit 64 Display unit 65 Operation unit 91 Network 93 Network 100 Excluded keyword table 200 Integrated keyword table 300 Data table 310 Integrated keyword 320 Log data 400 Integrated data table 410 Summary data 413 Integrated keyword 415 Data source 420 Extracted data 421a FAQ ID 421b Title information 423a Call log ID 423b Data generation date 423c Subject 425a Social log ID 425b Data generation date 425c Message information, 500... ratio data table, 510... integrated keyword, 520b... ratio, 520Bb... ratio, 530b... ratio, 540b... ratio, 600... correspondence priority data table , 605... keyword information, 620... data ratio, 630... call log data ratio, 640... social log data ratio, 650... response priority information, 700... problem solving data table , 705... keyword information, 720... data ratio, 730... call log data ratio, 740... social log data ratio, 750... problem information, 760... response priority, 770. ... solution information, 800 ... priority data table, 810 ... integrated keyword, 820 ... first priority, 830 ... second priority, 840 ... correspondence priority

Claims (23)

the computer
A first log data group including a plurality of first log data, a second log data group including a plurality of second log data different in type from the first log data, and a support content data group including a plurality of support content data and get
Extracting and integrating first log data and second log data including first keyword information corresponding to one support content data from the support content data group from the first log data group and the second log data group generate the data,
Extracting problematic keyword information from among a plurality of pieces of first keyword information included in the integrated data based on a numerical relationship between the support content data and at least one of the first log data and the second log data. including,
Information processing methods. The proportion of support content data corresponding to the problem keyword information in the support content data group, the proportion of first log data corresponding to the problem keyword information in the first log data group, and the second log data group generating problem information for the problem keyword information based on the relationship with the ratio of the second log data corresponding to the problem keyword information among
The information processing method according to claim 1 . The ratio of support content data corresponding to the problem keyword information in the support content data group, the ratio of first log data corresponding to the problem keyword information in the first log data group, and the ratio of the second log data group generating first priority information for the problem keyword information based on a relationship with a ratio of the second log data corresponding to the problem keyword information;
The information processing method according to claim 2. A ratio of support content data corresponding to the problem keyword information in the support content data group is lower than a first threshold, and a ratio of first log data corresponding to the problem keyword information in the first log data group is a second threshold. generating first priority information higher than other problem keyword information when higher than the threshold;
The information processing method according to claim 3. generating second priority information for the problem keyword information based on the number of data in the first log data for the problem keyword information;
generating third priority information for the problem keyword information based on the first priority information and the second priority information;
The information processing method according to claim 3. Among the plurality of first keyword information, excluding first keyword information that satisfies a predetermined condition in terms of a ratio of the first log data to the first log data group;
The information processing method according to claim 1 . updating the first keyword information based on trend information acquired at predetermined intervals;
The information processing method according to claim 1 . generating the first priority information based on at least one of trend information and user attribute information acquired every predetermined period;
The information processing method according to claim 3. Exclude first log data corresponding to second keyword information having a small degree of association with the first keyword information from the first log data group;
The information processing method according to claim 1 . updating the second keyword information based on trend information acquired at predetermined intervals;
The information processing method according to claim 9 . The first log data includes one of call log data, social log data, and chat log data, and the second log data includes the other one of call log data, social log data, and chat log data.
The information processing method according to any one of claims 1 to 10. A first log data group including a plurality of first log data, a second log data group including a plurality of second log data different in type from the first log data, and a support content data group including a plurality of support content data and get
Extracting and integrating first log data and second log data including first keyword information corresponding to one support content data from the support content data group from the first log data group and the second log data group generate the data,
Control for extracting problematic keyword information from among a plurality of pieces of first keyword information included in the integrated data based on a numerical relationship between the support content data and at least one of the first log data and the second log data. including the part
Information processing system. The control unit controls the proportion of support content data corresponding to the problem keyword information in the support content data group, the proportion of first log data corresponding to the problem keyword information in the first log data group, and the generating problem information for the problem keyword information based on a relationship with a ratio of second log data corresponding to the problem keyword information in a second log data group;
The information processing system according to claim 12. The control unit controls the proportion of support content data corresponding to the problem keyword information in the support content data group, the proportion of first log data corresponding to the problem keyword information in the first log data group, and generating first priority information for the problem keyword information based on a relationship with a ratio of second log data corresponding to the problem keyword information in the two log data groups;
The information processing system according to claim 13. The control unit controls a ratio of supporting content data corresponding to the problematic keyword information in the supporting content data group to be lower than a first threshold, and first log data corresponding to the problematic keyword information in the first log data group. When the ratio of is higher than the second threshold, generating first priority information higher than other problem keyword information,
The information processing system according to claim 14. The control unit generates second priority information for the problem keyword information based on the number of data in the first log data for the problem keyword information,
generating third priority information for the problem keyword information based on the first priority information and the second priority information;
The information processing system according to claim 14. The control unit excludes, from among the plurality of pieces of first keyword information, first keyword information that satisfies a predetermined condition in terms of a ratio of the first log data to the first log data group.
The information processing system according to claim 12. The control unit updates the first keyword information based on trend information acquired at predetermined intervals.
The information processing system according to claim 12. The control unit generates the first priority information based on at least one of trend information and user attribute information acquired every predetermined period,
The information processing system according to claim 14. The control unit excludes, from the first log data group, first log data corresponding to second keyword information having a low degree of association with the first keyword information.
The information processing system according to claim 12. The control unit updates the second keyword information based on trend information acquired at predetermined intervals.
The information processing system according to claim 20. wherein the first log data includes one of call log data, social log data, and chat log data, and the second log data includes the other of call log data, social log data, and chat log data; including one
The information processing system according to any one of claims 12 to 21. to the computer,
A first log data group including a plurality of first log data, a second log data group including a plurality of second log data different in type from the first log data, and a support content data group including a plurality of support content data and get
Extracting and integrating first log data and second log data including first keyword information corresponding to one support content data from the support content data group from the first log data group and the second log data group generate the data,
Extracting problematic keyword information from among a plurality of pieces of first keyword information included in the integrated data based on a numerical relationship between the support content data and at least one of the first log data and the second log data. program to run.

Images