JP2019204157A - Inquiry apparatus specification system, inquiry apparatus specification method - Google Patents

Abstract

To provide a technique for quickly identifying an apparatus to be an object of an inquiry.SOLUTION: An inquiry apparatus specification system has: a storage unit which stores a plurality of records in which apparatus identification information is associated with a plurality of pieces of property information related with the apparatus; a keyword extraction unit which extracts keywords from inquiry contents related with the apparatus; a data retrieval unit which retrieves the record including the keyword in values of the property information from the storage unit, specifies one or more pieces of property information not including the keyword in the value, when the plurality of records are retrieved, calculates a diversity for each specified property information, and determines the property information as an additional question item on the basis of a magnitude relation of each diversity; a screen generation unit which creates a screen which displays the determined property information, and displays a list of values included in a row of property information, in a form which a user can select; and a display unit which displays the screen.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for assisting input of inquiry contents for a device delivered to a customer and identifying the inquiry device.

  In general, the manufacturer of a device or the company in charge of maintenance of the device responds to customer inquiries about the device, and depending on the situation, dispatches maintenance personnel to the customer to investigate and repair the status of the device. And so on. Inquiries from customers are made, for example, by telephone calls to a call center, or by inputting into an inquiry web page prepared by the call center.

  In such an inquiry form, when the call center operator receives an inquiry, the call center operator grasps the keyword from the inquiry contents entered in the dialogue with the customer or the web page, and for example, the contract contents registered in the customer database etc. The target device is identified by searching.

  With regard to such an aspect, a technique related to a call center system that improves the efficiency of inquiries and can handle a high degree of customer satisfaction is disclosed (for example, Patent Document 1).

Japanese Patent Laid-Open No. 2006-171359

  When there are a large number of devices delivered to customers due to expansion of business formats and business details, keywords and management numbers are required as search keys to identify the devices. come. However, if the inquiring user does not know the information required to identify the target device, such as the facility name of the delivery destination, the customer management number that uniquely identifies the customer, and the keyword related to the device, the target device is identified. It takes a long time to identify the inquiry device quickly.

  In addition, if these keywords are unknown, this operation takes time because maintenance personnel need to be dispatched to the site to search for target devices.

  The present invention provides a technique for quickly identifying a device that is an object of an inquiry.

  In order to solve the above-mentioned problems, a typical inquiry device identification system is an inquiry device identification system that identifies a target inquiry device based on the inquiry content about a device input by a user, and uniquely identifies the device. A storage unit that stores a plurality of records in which device identification information and a plurality of attribute information related to the device are associated; a keyword extraction unit that extracts a keyword from the inquiry content related to the device input by the user; When a record including the keyword extracted by the keyword extraction unit in the attribute information value is searched from the storage unit and a plurality of records are searched, the plurality of records are processed, and the keyword is One or a plurality of attribute information not included in the value is specified, and each specified attribute information is included in the attribute information column. A data search unit for calculating the degree of dispersion of each value to be determined, and determining attribute information as an additional question item based on the magnitude relationship of each diversity, and the attribute information determined by the data search unit A screen generation unit for creating an additional question input screen that displays as an additional question item and displays a list of values included in the attribute information column in a format that can be selected by a user, and the screen generation unit And a display unit for displaying an additional question input screen created by.

It is possible to quickly identify the device that is the subject of the inquiry.
Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.

It is a figure which shows the example of whole structure of the inquiry apparatus specific system of 1st Embodiment. It is a figure which shows the hardware structural example of the inquiry apparatus specific device of embodiment. It is a flowchart which shows the operation example of embodiment. It is a figure which shows an example of the inquiry input screen and confirmation screen of embodiment. It is a figure which shows the data structural example used by embodiment. It is a figure for demonstrating the diversity of embodiment. It is a flowchart which shows the operation example when producing an additional question input screen. It is a figure which shows an example of the additional question input screen of 1st Embodiment. It is a figure which shows the example of whole structure of the inquiry apparatus specific system of 2nd Embodiment. It is a figure which shows a part of apparatus table of 2nd Embodiment, and is a figure which shows an example of a duplication degree. It is a figure which shows an example of the additional question input screen of 2nd Embodiment.

  In the present embodiment, a system configuration and a method for quickly identifying a variety of devices delivered to a customer when the customer makes an inquiry will be described. Further, in this embodiment, an elevator such as an elevator or a skater delivered to a customer will be described as a search target device.

  The system according to the present embodiment analyzes input information in order to identify a device that has been inquired by the customer, and displays an additional question for narrowing down to the customer. And the system of this embodiment receives the answer with respect to these additional questions from a customer, further narrows down an applicable apparatus and specifies an object apparatus.

  As a result, even when there are a large number of devices delivered to the customer and a wide variety of situations, the inquiry device can be quickly identified.

(First embodiment)
FIG. 1 is a block diagram illustrating a system configuration example of the first embodiment. The inquiry device specifying system 1000 includes a user terminal 1, an inquiry device specifying device 3, and a control terminal 4.

  The user terminal 1 is a general smartphone, tablet terminal, personal computer or the like used by a customer, and is a terminal connected to a network and capable of web access. An HMI (Human Machine Interface) 11 of the user terminal 1 is a display unit (web browser) that displays a screen for receiving input from a customer and displays search results. A dedicated client application that provides the same function may be used. The processing control unit 12 performs overall control of the user terminal 1 and is also responsible for data transfer between various applications and data transfer between various applications and various hardware.

  The position information acquisition unit 13 is a functional unit that acquires the current position of the user terminal 1, and is a functional unit that performs position measurement using radio waves from, for example, a GPS (Global Positioning System) or a wireless access point. . The position information acquisition unit 13 can also specify a rough position based on the IP address assigned to the user terminal 1. The position information acquisition unit 13 acquires the position when the user inputs by such implementation.

  The communication unit 14 is connected to the network 2 a and controls communication with the inquiry device specifying apparatus 3. The network 2a is a public IP network provided by a communication carrier in this embodiment.

  The inquiry device specifying device 3 is a device that receives an inquiry about a device (elevator, escalator, etc.) from the user terminal 1 and specifies the corresponding device from various devices delivered to the customer. In the present embodiment, the inquiry device specifying device 3 is incorporated in a control system that remotely monitors devices delivered to customers.

  The communication unit 31 of the inquiry device specifying apparatus 3 is connected to the network 2a and controls communication with the user terminal 1 and the like. The screen generation unit 37 has a function of a web server and generates a screen to be displayed on the HMI 11 of the user terminal 1. The screen generation unit 37 generates a screen written in HTML format or a script executed on the user terminal 1 side. A specific example of the screen created by the screen generation unit 37 will be described later.

  The text analysis unit 32 includes a morpheme analysis unit 32 a having a function of analyzing a morphological analysis (classifying words of a sentence based on part of speech) of a query sentence (text) input from the user terminal 1. Further, the text analysis unit 32 includes a keyword extraction unit 32b that extracts a keyword from the sentence after the morphological analysis using the keyword dictionary 35. A specific example of the keyword dictionary 35 will be described later.

  The processing control unit 30 performs overall control of the inquiry device specifying device 3 and is also responsible for data exchange between various applications and data exchange between various applications and various hardware.

  The temporary storage unit 36 is a functional unit for temporarily storing data when performing each operation described in detail below.

  The data search unit 33 is a functional unit that searches the control DB (database) 34 in which tables such as the property information table 34a and the device table 34b are constructed. The data search unit 33 also includes a diversity calculation unit 331 that operates when an additional question input screen is generated. Specific operations of the diversity calculation unit 331 will be described later. Specific examples of the property information table 34a and the device table 34b will be described later.

  The control terminal 4 is a terminal (personal computer) connected to the inquiry device specifying device 3 via the communication unit 43 or the local network 2b. The control terminal 4 is a terminal for mainly performing maintenance such as registration, deletion, and update of data for the property information table 34a, the device table 34b, and the keyword dictionary 35 of the inquiry device specifying device 3. For the HMI 41, the process control unit 42, and the communication unit 43 installed in the control terminal 4, the same functions as the HMI 11, the process control unit 12, and the communication unit 14 installed in the user terminal 1 are provided. I will omit the explanation here.

  FIG. 2 is a diagram illustrating a hardware configuration example of the inquiry device specifying apparatus 3 and an example of a computer configuration. Therefore, the user terminal 1 and the control terminal 4 also have the same hardware configuration.

  The inquiry device specifying apparatus 3 includes a controller 101, an input device 110, and an output device 111. The controller 101 controls each piece of hardware that operates inside the inquiry device specifying apparatus 3. The controller 101 has the following configuration.

  A CPU 102 (CPU: Central Processing Unit) is a processing device that develops a program stored in a ROM 104 (ROM: Read only memory) or a storage 105 in a RAM 103 (RAM: Random access memory) and executes an operation. The CPU 102 performs overall control of the hardware in the controller 101 by executing a program. The RAM 103 is a volatile memory, and is a work memory when the CPU 102 performs processing. The RAM 103 temporarily stores necessary data while the CPU 102 is executing a program.

  The ROM 104 is a non-volatile memory, and stores a basic input / output system (BIOS) that is executed by the CPU 102 when the inquiry device specifying apparatus 3 is activated, and firmware.

  The storage 105 is an auxiliary storage device that stores data in a nonvolatile manner, and is a hard disk drive (HDD), a solid state drive (SSD), a large-capacity flash memory, or the like. The storage 105 stores programs executed by the CPU 102 and control data. In the present embodiment, programs and databases that constitute each functional unit shown in FIG. 1 are introduced and built in the storage 105 in advance.

  The network I / F 106 is an interface board that controls data communication performed with an external device.

  The input I / F 107 is an interface that controls input / output of signals with the input device 110. The output I / F 108 receives an instruction from the CPU 102 and causes the output device 111 to draw a display image or video.

  The input device 110 is a keyboard or a mouse, for example, and the output device 111 is a flat monitor or a display. When the configuration shown in FIG. 2 is applied to the user terminal 1 or the control terminal 4, the input device 110 and the output device 111 may be replaced with a touch panel display. In addition to the configuration of FIG. 2, the user terminal 1 may include a device or device that can acquire position information, such as a GPS receiving unit.

  The hardware configuration shown in FIG. 2 and the program installed in the storage 105 operate in cooperation to realize each functional unit shown in FIG. In the present embodiment, it is assumed that the operation is performed by software control as described above. However, some functions may be provided by using an integrated circuit. Thus, even a configuration including an integrated circuit can be treated as one aspect of a computer.

  Next, the specific operation of the inquiry device of this embodiment will be described using the flowchart of FIG. Here, the flowchart of FIG. 3 explains each of the operation in the user terminal 1 and the operation in the inquiry device specifying apparatus 3. Communication occurring between the user terminal 1 and the inquiry device specifying device 3 is indicated by broken-line arrows between the flowcharts.

  First, when a customer makes an inquiry about, for example, an elevator problem, the customer terminal 1 activates the HMI 11 and accesses the inquiry device specifying device 3. Then, the HMI 11 displays the inquiry input screen generated by the screen generation unit 37 (S001). An example of this inquiry input screen is shown in FIG. The inquiry input screen 400 includes an example field 401 that displays an input example, an editable input field 402 that accepts input of inquiry contents from a customer, and a send button 403.

  Further, when the inquiry input screen 400 is displayed by the HMI 11, the position information acquisition unit 13 performs an acquisition process of the current position of the user terminal 1 (S002). The position information acquisition unit 13 attempts to acquire a positioning value (latitude, longitude) by GPS or the like. If the position information can be acquired (S003: Yes), the process proceeds to S005. On the other hand, when the position information cannot be acquired (S003: No), the HMI 11 displays a screen for specifying the position on the address input field or the map, and requests the customer to input the position information (S004). Note that the process of S004 may be realized by cooperating with another map information system or using a conventional technique.

  The communication unit 14 acquires the position information obtained by the position information acquisition unit 13 and the position information obtained by the HMI 11 via the processing control unit 12 and transmits them to the inquiry device specifying apparatus 3 (S005). In addition, it is desirable to obtain customer consent for the acquisition of the position information. Therefore, in this embodiment, the HMI 11 displays the confirmation screen 410 shown in FIG. If the consent button 411 is pressed here, the communication unit 14 transmits position information. If the disagreement button 412 is pressed, the transmission of position information is cancelled.

  The customer uses the inquiry input screen 400 displayed by the HMI 11 to input the state of the inquiry device (elevator) and the inquiry content into the input field 402 while referring to the example shown in the example field 401. The HMI 11 accepts this input (S006).

  After the input, when the customer presses the send button 403 in the inquiry input screen 400, the HMI 11 delivers the text data (referred to as inquiry text) in the input field 402 to the communication unit 14 via the processing control unit 12. . And the communication part 14 transmits this inquiry text toward the inquiry apparatus specific apparatus 3 (S007).

  Subsequently, the operation of the inquiry device specifying apparatus 3 will be described. The communication unit 31 of the inquiry device specifying device 3 always waits for reception of position information from the user terminal 1 and reception of an inquiry text. When receiving the position information from the user terminal 1 (S101), the communication unit 31 delivers the position information to the data search unit 33 via the processing control unit 30.

  The data search unit 33 refers to the property information table 34a using the acquired location information, and narrows down the target property (facility such as a building or a condominium) in which the device is located, located in the vicinity of the location information (S102). With this operation, the inquiry device specifying device 3 can narrow down the target properties around the customer who uses the user terminal 1 within an approximate range.

  FIG. 5A is a diagram illustrating a data configuration example of the property information table 34a. In the property information table 34a, each building code that uniquely identifies a property is associated with a business office or sales office that has jurisdiction over the property, a management number, a building name, a building address, and location information (latitude, longitude). Registered. The position information registered in the property information table 34a is information indicating the position of the facility where the delivered equipment is installed. The data search unit 33 searches the property information table 34a for a record including the position information (latitude, longitude) that is numerically closest to the position information (latitude, longitude) obtained in S101, and obtains the building code. Alternatively, the data search unit 33 ranks the records in the property information table 34a so as to be in the order from the position information obtained in S101, selects the top several items as candidates, and selects these buildings. Get the code. In this way, the data search unit 33 narrows down the surrounding target properties.

  If location information is not transmitted because the disagreement button 412 has been pressed or the like, the processing of S102 is not performed, and the search is limited only to the inquiry text described below.

  Next, when the communication unit 31 receives the inquiry text from the user terminal 1, it is delivered to the text analysis unit 32 via the processing control unit 30. The text analysis unit 32 generates a keyword from the delivered inquiry text (S103).

  In S103, the morpheme analysis unit 32a of the text analysis unit 32 performs morphological analysis on the query text, thereby decomposing the query text into words. Then, the keyword extraction unit 32b extracts a keyword from the query text using the text data after the morphological analysis and the keyword dictionary 35.

  The operation of S103 will be described with a specific example. When the inquiry text is, for example, “There was an unusual sound from the EV of the apartment in front of the convenience store this morning”, the morpheme analysis unit 32a, for example, “Today | | | Morning |, | | | | | | | | | | | | | | ing).

  Next, the keyword extraction unit 32b creates an appropriate keyword from the analysis result of the morpheme analysis unit 32a. Whether or not the keyword is appropriate is determined by whether or not it is registered in the keyword dictionary 35.

  FIG. 5B is a diagram illustrating a part of the keyword dictionary 35. The keyword dictionary 35 has a configuration in which input information and keywords are associated with each other, and each record is registered in advance. As described above, when “EV” is included in the inquiry text, the keyword extraction unit 32b searches the keyword dictionary 35 for a record including the “EV” in the input information. Then, the keyword extraction unit 32b extracts “elevator” associated with “EV” as a keyword after replacement. As another example, the keyword extraction unit 32b can obtain “elevator” that is a keyword after replacement by searching for a record including “elevator” in the input information. As described above, the keyword extraction unit 32b can also create a keyword by converting an abbreviation into a regular name, or can create a keyword in which notation fluctuation is eliminated.

  The keyword extraction unit 32b searches for all the words obtained by the morphological analysis unit 32a in a brute force manner and creates each keyword. A word for which a search result cannot be obtained is a word that cannot be an appropriate keyword and cannot be used in subsequent processing, and thus is not considered in subsequent processing. For the purpose of improving the quality of search, history data indicating that a word other than an appropriate keyword has been searched may be left. By such processing of the keyword extracting unit 32b, it is possible to obtain appropriate keywords (keywords in which abbreviations and notation fluctuations are eliminated) suitable for subsequent processing.

  The text analysis unit 32 delivers the obtained keyword to the data search unit 33. Further, the text analysis unit 32 stores the obtained keyword and inquiry text in the temporary storage unit 36.

  Next, the data search unit 33 searches the device table 34b using the building code extracted in S102 and the keyword acquired from the text analysis unit 32, and narrows down target devices (S104). Here, when a plurality of keywords are obtained from the text analysis unit 32, a search is performed for all the keywords.

  FIG. 5C is a diagram illustrating an example of the device table 34b. The device table 34b has a data configuration that stores a plurality of records in which device numbers (device identification information) for uniquely identifying devices and attribute information related to the devices are associated with each other. In the present embodiment, the building code, the device type (“elevator”, “escalator”) indicating the type and classification of the device, the usage of the device (“passenger”, “emergency”, “for carrying luggage”), Unit 1 No. 2 and No. 2, etc., which are management numbers of elevators attached in the facility, and special notes are attribute information. The contents of the special notes are information obtained from a maintenance person or a customer.

  The data search unit 33 searches for a record in which the value of the building code extracted in S102 is included in the building code column of the device table 34b, and includes the keyword generated in S103 in each attribute information column. To extract records. The operation of specifying the inquiry device in this embodiment is an operation of finally specifying one record from the device table 34b and obtaining a device number by performing such extraction processing and the like.

  The data search unit 33 determines whether a device is specified by the operation of S104, that is, whether only one record is extracted from the device table 34b (S105).

  When only one record is extracted (S105: Yes), the data search unit 33 includes the data in the record specified as one from the device table 34b, and the records narrowed down from the property information table 34a in S102. Are stored in the temporary storage unit 36 in association with each other.

  The screen generation unit 37 incorporates various types of specified data stored in the temporary storage unit 36 to generate an HTML display screen. The communication unit 31 transmits this display screen to the user terminal 1 as a specific result (S109). The communication unit 14 of the user terminal 1 receives this display screen, and the HMI 11 displays this display screen (S010).

  Note that data such as the specified device number may be transmitted to a linked monitoring system. In this case, the monitoring system performs an operation of acquiring operation data for the device associated with the received device number, and analyzes the state of the failure based on the operation data.

  On the other hand, as a result of the search in S104, if a plurality of records are searched without being narrowed down to one (S105: No), the data search unit 33 generates an additional question for further narrowing down the devices (S106). ).

  A method for generating additional questions will be described below. Here, a case where the bill code “1234-A567” is obtained in S102 and the keyword “elevator” is obtained in S104 will be described as an example.

  First, the data search unit 33 extracts a record whose bill code is “1234-A567” from the device table 34b shown in FIG. By this operation, a total of seven records from the top record shown in FIG. 5C to the seventh row record are extracted. Then, the data search unit 33 extracts a record including the keyword “elevator” in the column from the seven records. Note that the column names of the device type, usage, machine, and special notes are uniquely determined, so the data search unit 33 creates an SQL statement that specifies each column name in the where clause and keywords in the like clause. Then, a record is extracted using this SQL sentence. Here, since the character string “Elevator” is included in the device type column from the top record of FIG. 5C to the sixth record, these six records are further narrowed down. Extracted. That is, at this stage, the data search unit 33 can obtain a total of six records as inquiry device candidates.

  FIG. 6 is a diagram showing six records narrowed down by the above procedure, and is a diagram illustrating the diversity of the present embodiment.

  The diversity calculation unit 331 of the data search unit 33 extracts a column (column) that has not been used for the search so far, that is, a column (column) that does not include a keyword in the processing so far. In the example of FIG. 6, each column of usage, number machine, and special notes is taken out. Next, the diversity calculator 331 calculates the diversity for each extracted column.

  Here, the diversity is the number of cases after the duplication of data in the column is removed. In the example of FIG. 6, the data in the usage column can be classified into “passenger”, “emergency”, and “cargo transportation”. The number 3 is the number of cases after duplication of data is eliminated. It becomes diversity. Similarly, since the data of the machine column is six different data “1”, “2”, “3”, “4”, “5”, and “6”, the diversity is 6. The special item column also has a diversity of 6, since each data is different. The diversity calculation unit 331 calculates the diversity for each column by using an SQL statement that incorporates a COUNT phrase and DIRTINCT.

  The diversity calculation unit 331 compares the calculated degrees of diversity in each column, and identifies the column with the largest diversity. This corresponds to the machine column and the special notes column. The data search unit 33 creates a list of each column by removing duplicate values for each column (the machine column and the special note column) specified by the diversity calculation unit 331. This list is a display item of a drop-down list described later. The data search unit 33 delivers the created list and each data stored in the temporary storage unit 36 to the screen generation unit 37.

  The screen generation unit 37 incorporates various data acquired from the data search unit 33 and generates an additional question input screen in HTML format. Up to here is the operation of S106.

  FIG. 7 is a flowchart showing the operation of S106. The data search unit 33 extracts the search results (records) narrowed so far from the device table 34b (S201). The diversity calculation unit 331 performs the following processes of S202 and S203 on the plurality of records extracted in S201 as processing targets. That is, the diversity calculation unit 331 identifies one or more columns that do not include the keyword in the value (S202), and calculates the diversity for each identified column (S203).

  Then, the data search unit 33 determines a column to be an additional question item based on the magnitude relationship of diversity (S204). In the present embodiment, the data search unit 33 determines the column having the greatest diversity as the additional question item, and creates a list. The screen generation unit 37 displays each value in the column row determined by the data search unit 33 (that is, the list created in S203) in a format that can be selected by the customer customer. A screen is created (S205).

  After the processing shown in FIG. 7 is performed, the communication unit 31 transmits an additional question input screen toward the user terminal 1 (S107 in FIG. 3). The communication unit 14 of the user terminal 1 receives the additional question input screen, and the HMI 11 displays it (S008).

  FIG. 8A is a diagram illustrating an example of an additional question input screen displayed on the HMI 11. The additional question input screen 500 includes items 502 for specifying an inquiry device from a customer. Here, since the building name, address, and device type are specified, black shading is displayed, and unconfirmed items (additional question items) of the unit and special notes are displayed in white. For the fields that are outlined, a drop-down list is displayed by operating the buttons 502a and 502b.

  FIG. 8B shows a drop-down list 510 for the machine that is displayed when the button 502a is operated, and FIG. 8C is for special instructions that are displayed when the button 502b is operated. A drop-down list 520 is shown. Each of these drop-down lists is a list of data for additional questions specified by the data search unit 33 (see the machine number column and the special note column in FIG. 6).

  The HMI 11 receives an answer to the additional question selected from the drop-down list (S009). When the transmission button 503 shown in FIG. 8A is pressed, the communication unit 14 transmits the response data to the inquiry device specifying apparatus 3 via the processing control unit 12. The answer data is, for example, the text itself of the answer item selected from the list (such as “back right when viewed from the front entrance”) or information uniquely identifying the answer item.

  The communication unit 31 of the inquiry device specifying apparatus 3 receives the response data and delivers it to the data search unit 33. Upon obtaining this answer data, the data search unit 33 returns the process to S105 (see FIG. 3) and further narrows down including the answer, thereby re-specifying the inquiry target device (S105). If it is possible to specify one here (S105: Yes), the process proceeds to S109. On the other hand, when it cannot be specified yet (S105: No), additional question generation processing is performed again (S106).

  In the case of the example described so far, one record is specified from the device table 34b in S105 at this stage. If it is not specified, an additional question input screen for inquiring each data (“passenger”, “emergency”, “cargo transportation”) in the application column that has not been used so far is generated in S106. Is done. If there are multiple columns that have not been used so far, even at this stage, as in the above case, calculate the diversity of each column that was not used and compare the levels of diversity to determine the columns for additional questions. The specified operation is performed.

  Note that the data search unit 33 may determine the attribute information to be the additional question item so that at least the attribute information having the greatest diversity is included. As an example of this, there is an implementation example in which a column having a degree of diversity equal to or greater than a prescribed threshold is an additional question item, in addition to the above implementation example in which only the column with the largest diversity is an additional question item.

  According to the aspect of the above embodiment, when there is an inquiry from a customer about the device delivered to the customer, an appropriate additional question can be presented even if the inquiry device cannot be specified only by the information presented by the customer. it can. Therefore, it is possible to quickly identify the inquiry device.

(Second Embodiment)
In the second embodiment, another example of creating an additional question will be described.

  FIG. 9 is a block diagram illustrating a system configuration example of the second embodiment. The data search unit 33 of the second embodiment further includes a duplication degree calculation unit 332. Other system configurations, device configurations, operation examples, and creation of additional questions using diversity are the same as those described with reference to FIGS. 1 to 7 of the first embodiment. To do.

  In the second embodiment, the property information table 34a shown in FIG. 5A and the keyword dictionary 35 shown in FIG. 5B are the same as those in the first embodiment, but the device table 34b is explained. For convenience, use a different numbering system.

  FIG. 10A is an excerpt of the device table 34b in the second embodiment, which corresponds to FIG. 6 of the first embodiment. In the second embodiment, a passenger elevator and an emergency elevator for mainly transporting people and a luggage transport elevator for mainly transporting luggage, and separately managing the number of each unit number (Refer to the machine column in FIG. 10A).

  In the first embodiment, the additional question input screen is generated so that only the column with the largest diversity is an additional question. However, in the second embodiment, a screen for asking questions about all the columns is generated. To do. Further, when generating the additional question input screen, the screen generation unit 37 generates the input screen so that the additional item fields are arranged in descending order of diversity.

  In the example of FIG. 10A, the diversity of the application column and the diversity of the special mention column are diversity: 3 and diversity: 6, respectively, as in the first embodiment. The number column can be classified into four data of “1”, “2”, “3”, and “4”, and therefore the diversity is 4. Therefore, in the case of this example, the screen generation unit 37 creates an additional question input screen in the order of special notes, machine numbers, and usages from the top according to the order of diversity (see FIG. 11A described later).

  In addition, in the second embodiment, a list of drop-down lists is displayed from the top in descending order of the number of cases. The degree of duplication is calculated for each data so that the display order is as described above. The degree of overlap will be described with reference to FIG. FIG. 10B shows the degree of duplication of each piece of data from the table shown in FIG. In the present embodiment, the degree of duplication is a value indicating the number of registered data items in a column. For example, if the usage column in FIG. 10A is the target column, three “passenger” data, one “emergency” data, and two “cargo transportation” data are registered in the usage column. Has been. Therefore, the degree of duplication of “riding” is 3, the degree of duplication of “emergency” is 1, and the degree of duplication of “for luggage” is 2. Similarly, if the machine column is the target column, “1” machine has a duplication degree of 2, “2” machine has a duplication degree of 2, “3” machine has a duplication degree of 1, and “4” machine has a duplication degree of 1. It becomes. Since the special note column is all different data, the degree of duplication of each data is 1.

  The duplication degree calculation unit 332 of the data search unit 33 calculates the duplication degree for each data in the attention column by using an SQL statement incorporating COUNT or GROUP BY, and the attention column is arranged in the order of the duplication degree. Sort each data in to create one-dimensional array data. For example, in the case of a usage column, the degree-of-duplication calculation unit 332 generates one-dimensional array data sorted in the order of [riding, luggage transport, emergency]. The duplication degree calculation unit 332 performs this process for each column, and thereafter associates the column with the one-dimensional array data. The redundancy calculation unit 332 adds, for example, a column name to [Ride, Cargo Transport, Emergency] and adds array data such as [Use, Passenger, Cargo Transport, Emergency] (herein referred to as column array data). By associating the column name with the one-dimensional array data. The redundancy calculation unit 332 generates column array data such as [Unit, 1, 2, 3, 4] for the machine column, and similarly generates column array data for the special mention column.

  Then, the data search unit 33 further arranges the column arrangement data obtained by the duplication degree calculation unit 332 in order of diversity, and delivers this to the screen generation unit 37. The screen generation unit 37 incorporates the acquired array data, creates an additional question input screen, and transmits it to the user terminal 1. The communication unit 14 of the user terminal 1 receives this additional question input screen, and the HMI 11 displays it.

  FIG. 11A is a diagram illustrating an example of an additional question input screen displayed on the HMI 11 of the second embodiment. In the additional question input screen 500 of the second embodiment, the white columns of the items 502 are displayed in the order of the special notes (diversity: 6), number (diversity: 4), and usage (diversity: 3). It is arranged in the order of diversity.

  FIG. 11 (B) is a drop-down list 510 for the machine displayed when the button 502a in the second embodiment is operated, and FIG. 11 (C) is displayed when the button 502c is operated. This is an application drop-down list 530. In either case, the list is displayed in descending order of redundancy. The drop-down list of special items has the same display contents as in the first embodiment (see FIG. 8C).

  Here, the diversity described in the first and second embodiments will be further described. Diversity is the number of cases after duplication of data in the relevant column as described above. The greater the diversity, the greater the diversity of data, and vice versa. This shows that the same kind of data is collected. Diversity having such characteristics can be treated as an index value indicating the degree of data dispersion. In the first and second embodiments, as described above, the number of cases after the duplication of data in the relevant column has been treated as diversity, but another index indicating the degree to which the data is scattered, For example, variance or standard deviation may be used as the diversity. In addition, by adopting the number of cases after the duplication of data in the relevant column as diversity, it is possible to obtain the degree of dispersion of each value without performing complicated arithmetic processing such as dispersion and standard deviation. it can.

  Further, as in the first and second embodiments, by presenting an item with a high degree of diversity as an additional question, the customer selects one from a wide variety of data. In this way, the inquiry device can be considerably narrowed down by having the customer select one of a wide variety of data. Therefore, the number of steps until the inquiry device is specified can be reduced by making an item with a large variety as an additional question.

  In addition, as can be seen from the display order of various search sites, the item displayed at the top has a higher probability of being watched than the item displayed at the bottom. As in the second embodiment, by displaying items with a high degree of diversity in order from the top of the screen, the probability that the item with the high degree of diversity is watched and selected increases. Such a display order is also helpful in reducing the number of steps.

  In the second embodiment, the mode of displaying a list of drop-down lists in the order of items having a high degree of overlap, that is, in the order of the number of registered cases has been described. Since data with a large number of registrations is data that is often selected by the user, displaying this data at the top of the list can reduce the trouble of searching for an answer item and improve convenience. it can.

  You may combine 1st Embodiment and 2nd Embodiment as needed.

  In the first and second embodiments, an inquiry is received, a search result is displayed, an additional question is displayed / selected, and the like by a browser or an application on the user terminal. The mode is not limited to this method, and input and output by telephone voice may be performed using a call analysis or voice analysis technique.

  In the first and second embodiments, the system configuration in which the inquiry device specifying device 3 is a server and the user terminal 1 is a client has been described. However, by incorporating a functional unit equivalent to the HMI 11 into the inquiry device specifying device 3, It is good also as a system structure only of the inquiry apparatus specific apparatus 3. FIG. In this case, the user who operates and operates the system is not the customer but the operator, and the operator inputs the inquiry contents by telephone from the customer. In this case, the location information may be obtained by obtaining an address from the customer via a telephone or the like.

  In the present embodiment, the drop-down list is adopted as a format for displaying a list of answer values. However, for example, a format in which a list of answers is displayed in a pop-up may be used. In other words, as long as the list display for answers can be selected by the user, other display formats may be used.

  In addition, this invention is applicable not only to the aspect of said each embodiment but the inquiry regarding what kind of apparatuses, facilities, and products, such as office equipment and household appliances.

  As described in detail above, according to the present embodiment, it is possible to quickly identify the device that is the target of the inquiry.

  In addition, this invention is not limited to above-described embodiment, Various modifications are included. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to one having all the configurations described. Further, a part of the configuration of an embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of an embodiment. In addition, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment. Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function can be stored in a recording device such as a memory, a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

1: User terminal 3: Inquiry device identification device 11: HMI
12: Processing control unit 13: Position information acquisition unit 14: Communication unit 30: Processing control unit 31: Communication unit 32: Text analysis unit 32a: Morphological analysis unit 32b: Keyword extraction unit 33: Data search unit 34: Control DB
34a: Property information table 34b: Device table 35: Keyword dictionary 36: Temporary storage unit 42: Processing control unit 43: Communication unit 331: Diversity calculation unit 332: Duplication degree calculation unit 1000: Inquiry device identification system

Claims (7)

An inquiry device identification system that identifies a target inquiry device based on the inquiry content about the device input by a user,
A storage unit for storing a plurality of records in which device identification information for uniquely identifying a device and a plurality of attribute information related to the device are associated;
A keyword extraction unit that extracts keywords from the content of inquiries about devices input by users;
When a record including the keyword extracted by the keyword extraction unit in the value of attribute information is searched from the storage unit and a plurality of records are searched, the plurality of records are processed, and the keyword is One or a plurality of attribute information not included in the value is identified, and for each of the identified attribute information, a diversity indicating the degree of dispersion of each value included in the attribute information column is calculated, A data search unit that determines attribute information to be added question items based on the magnitude relationship,
An additional question input screen for displaying attribute information determined by the data search unit as an additional question item and displaying a list of values included in the attribute information column in a format that a user can select. A screen generator to be created;
A display unit for displaying an additional question input screen created by the screen generation unit;
Inquiry device identification system. The inquiry device identification system according to claim 1,
The inquiry device specifying system, wherein the data search unit determines attribute information to be an additional question item so as to include at least attribute information having the largest diversity. The inquiry device identification system according to claim 1,
Furthermore, it has a position information acquisition unit that acquires the position of the user when the user inputs,
The storage unit further stores location information of a facility where the device is installed in association with the device identification information,
The data search unit further narrows down records stored in the storage unit using the position acquired by the position information acquisition unit. The inquiry device identification system according to claim 3,
The inquiry device specifying system, wherein the data search unit calculates the number of cases after the duplication of data in the column is removed as the diversity. The inquiry device identification system according to claim 1,
The screen generation unit creates an additional question input screen by arranging the additional question items in order from the top in the order of the degree of diversity. The inquiry device identification system according to claim 1,
The inquiry device specifying system, wherein the screen generation unit generates an additional question input screen that displays a list of values included in the attribute information column in descending order of number of cases. From the inquiry content related to the device input by the user using a system having a storage unit that stores a plurality of records in which device identification information uniquely identifying the device and a plurality of attribute information related to the device are associated Inquiry device identification method for identifying the target inquiry device,
The system
Extract keywords from inquiries about devices entered by users,
Search the storage unit for a record that includes the extracted keyword in the value of the attribute information,
When a plurality of records are searched, one or a plurality of attribute information that does not include the keyword in the value is specified for the plurality of records, and for each of the specified attribute information, Calculate the diversity that indicates the degree of dispersion of each value in the column,
Based on the magnitude relationship of each diversity, determine the attribute information that will be the additional question items,
Display the determined attribute information as an additional question item, and create an additional question input screen that displays a list of values included in the attribute information column in a format that the user can select,
Display the created additional question input screen,
Inquiry device identification method characterized by that.

Images