JP2007140841A - Information processor and its control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform more reliable Internet examination by dealing with qualitative information and quantitative information easily and comprehensively. <P>SOLUTION: The information processor 106 generates an element table 600 according to the qualitative information and quantitative information included in answer information obtained from a questionnaire. Then, the information processor 106 uses an element designated by an item 708 in an order form 700 to cluster answerers. Next, the information processor 106 provides an orderer with an analysis report 1000 which includes property information on each cluster and information on ratio. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an information processing apparatus that analyzes questionnaire response information and a control method thereof.

  Today, survey methods using the Internet (hereinafter referred to as “Internet surveys”) are often used as a means of obtaining marketing information such as customer satisfaction (CS) and brand evaluation indexes for products and services. . According to the Internet survey, respondents can easily respond to questionnaires, so that there is an advantage that a large amount of information (answers) can be collected quickly and at low cost. Further, the Internet survey is not limited to the CS survey and the brand evaluation index survey, and can be widely applied to general market surveys.

  In such a questionnaire collection method based on the Internet survey, two types of information, that is, qualitative information and quantitative information can be acquired. Here, the qualitative information typically refers to information obtained by a free description type answer. The quantitative information typically means information obtained by answering options. In addition, a question for obtaining qualitative information as an answer is called a qualitative question, and a question for obtaining quantitative information is called a quantitative question.

Further, Patent Document 1 discloses a method of collecting questionnaire response information via the Internet, summing up and analyzing the response information with a server, and presenting an analysis report to the client via the Internet.
JP 2004-348665 A

  Thus, conventionally, qualitative information and quantitative information acquired in Internet surveys are individually handled, and they cannot be easily and integratedly handled. Therefore, it is not possible to generate an Internet survey analysis report (CS information or the like) with higher reliability.

  Moreover, although the method disclosed in Patent Document 1 is a kind of Internet survey, it does not consider how to handle qualitative information and quantitative information in the analysis of response information.

  Therefore, an object of the present invention is to enable more reliable Internet surveys by easily and integratively handling qualitative information and quantitative information.

  In order to solve the above problems, an information processing apparatus of the present invention is an information processing apparatus connected to a plurality of client terminals via the Internet, from a plurality of respondents who are users of the plurality of client terminals. Acquisition means for acquiring first qualitative information and first quantitative information via the plurality of client terminals as answer information to a qualitative question and a quantitative question, and the first qualitative information in units of words Decomposing means for decomposing and generating second qualitative information, first selecting means for selecting a predetermined question from the qualitative question and the quantitative question, and the first corresponding to the selected qualitative question A second selection means for selecting a predetermined word as the third qualitative information among the two qualitative information, the second quantitative information corresponding to the quantitative question selected from the quantitative information, Based of the and qualitative information, characterized in that it and a clustering means for classifying the plurality of respondents into a plurality of clusters.

  The information processing apparatus control method according to the present invention is a method for controlling an information processing apparatus connected to a plurality of client terminals via the Internet, from a plurality of respondents who are users of the plurality of client terminals. As the answer information to the qualitative question and the quantitative question, an acquisition step of acquiring the first qualitative information and the first quantitative information via the plurality of client terminals, and the first qualitative information in units of words A decomposing step of decomposing to generate second qualitative information; a first selecting step of selecting a predetermined question from the qualitative question and the quantitative question; and the first corresponding to the selected qualitative question A second selection step of selecting a predetermined word as the third qualitative information among the two qualitative information, second quantitative information corresponding to the quantitative question selected from the quantitative information, and the first Based of the and qualitative information, characterized in that it and a clustering step for classifying the plurality of respondents into a plurality of clusters.

  Other features of the present invention will become more apparent from the accompanying drawings and the following description of the best mode for carrying out the invention.

  According to the present invention, qualitative information and quantitative information can be easily and integratedly handled, and a more reliable Internet survey can be performed.

  Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.

<System configuration>
FIG. 1 is a diagram showing an outline of the configuration of a system used in the Internet survey of this embodiment.

  In the present embodiment, devices described below can communicate with each other via the Internet 100.

  The mobile terminal 102 is a terminal possessed by a respondent in the Internet survey. For example, a mobile phone having an Internet communication function for accessing the Internet 100, a PDA, and the like correspond to this. In addition, it is preferable that the respondent is a customer of a store that is a target of CS evaluation or the like. Hereinafter, the respondent may be referred to as a customer.

  A PC (personal computer) 104 is a PC that can be used by respondents to access the Internet 100. The respondent can answer the questionnaire by connecting to the information processing apparatus 106 described later via the Internet 100 using the mobile terminal 102 or the PC 104.

  The information processing apparatus 106 provides a questionnaire response web page (questionnaire answer form) to customers who have accessed via the Internet 100 using the mobile terminal 102 or the PC 104, and collects questionnaire answer results. In addition, the collected questionnaire response results are analyzed, and an analysis report is provided to the orderer (a person who uses the analysis report of the Internet survey). The information processing apparatus 106 can also instruct the POS terminal 110 of a questionnaire collection method. The information processing apparatus 106 has various other functions, but details will be described later with reference to FIGS. Note that the information processing apparatus 106 corresponding to the present embodiment is realized by, for example, a PC or a workstation (WS). Furthermore, the information processing device 106 may be configured from a single device, or may be configured so that a plurality of devices operate in cooperation.

  The PC 108 is a PC that can be used by the orderer. The PC 108 can receive the analysis result of the questionnaire from the information processing apparatus 106 via the Internet 100 in response to the orderer's instruction. The PC 108 can also instruct the POS terminal 110 on the questionnaire collection method via the information processing apparatus 106.

  The POS terminal 110 is a POS (point of sales) compatible register installed in a store (hereinafter referred to as “investigation store”) targeted for CS evaluation or the like. The POS terminal 110 can request a predetermined customer to answer a questionnaire in accordance with an instruction from the information processing apparatus 106. The response request can be realized by, for example, issuing a receipt in which the URL of the questionnaire response form (hereinafter referred to as “questionnaire request”) is written by the POS terminal 110.

  When the information processing apparatus 106 is viewed as a server apparatus, the mobile terminal 102, the PC 104, the PC 108, and the POS terminal 110 correspond to client terminals.

<Hardware Configuration of Information Processing Device 106>
FIG. 2 is a functional block diagram showing a hardware configuration of the information processing apparatus 106 corresponding to the present embodiment.

  The CPU 200 controls the entire information processing apparatus 106 by executing an application program, an operating system (OS), a control program, and the like stored in a hard disk drive (HDD) 205 described later.

  A ROM 201 is a non-volatile memory and stores programs such as a basic I / O program, various data such as font data and template data used for document processing.

  A RAM 202 is a volatile memory for temporarily storing various data, and functions as a main memory, a work area, and the like of the CPU 200.

  The external storage drive 203 is for realizing access to a recording medium (media) 204, and can load a program, data, and the like stored in the medium 204 into the information processing apparatus 106. As the media 204, for example, a floppy (registered trademark) disk, CD-ROM, CD-R, CD-RW, PC card, DVD, IC memory card, MO, or the like can be used.

  The HDD 205 stores an application program, an OS, a control program, a user dictionary, and the like corresponding to the embodiment of the present invention.

The user dictionary is a dictionary used to determine which words included in the qualitative answer are to be analyzed when the questionnaire answer is analyzed. The user dictionary excludes particles such as “ga” and “ha” included in the qualitative answer, nouns having no meaning, and the like from the analysis target. Since which word is subject to analysis differs depending on the research field, a different user dictionary is stored for each research field (for example, a CS survey at a convenience store and a CS survey at a gas station). Note that the user dictionary is generated in advance based on heuristics. Specifically, the information processing device is generated by executing the following procedures (1) to (4).
(1) Collect many questionnaire responses in the survey field.
(2) The qualitative answers included in the questionnaire answers are shared and the obtained words and keywords (including phrases) are extracted, synthesized, processed, and edited.
(3) Observe and extract which words and phrases are used in qualitative answers using the concordance function.
(4) Perform processing such as replacement / deletion while observing the frequency distribution of the words / phrases that have been divided.
(5) Observe (compare) the history of extraction / composition / processing / editing / replacement / deletion processing in (2) to (4) above, and finally register it as a user dictionary used for analysis.

  Examples of application programs include those listed below.

  1. Notifying the POS terminal 110 of the criteria (hereinafter referred to as “sampling conditions”) on which customer's receipt the questionnaire request is to be printed.

  2. Receives information on a receipt with a questionnaire request actually issued from the POS terminal 110 and calculates, for example, a questionnaire collection rate.

  3. An order for an Internet survey request from PC. The Internet survey request includes information on what kind of report is generated from the questionnaire response. Details of the Internet survey request will be described later with reference to FIG.

  4). This is to check whether the answer to the questionnaire received from the mobile terminal 102 or the PC 104 is an illegal answer such as a double answer.

  5. Analyzing questionnaire responses received from mobile terminal 102 or PC 104 and generating a report based on an Internet survey request.

  Note that each program, data, and the like stored in the HDD 205 may be stored in the medium 204.

  The instruction input device 206 includes a keyboard, a pointing device (such as a mouse), and the like. Using the instruction input device 206, the operator of the information processing device 106 can input a command or the like for controlling the information processing device 106.

  A display 207 displays a command input from the instruction input device 206, a response output of the information processing device 106 in response thereto, and the like.

  The interface (I / F) 208 is configured by a network I / F or the like, and is connected to the Internet 100 via this to exchange data with the PC 108 or the like.

  A system bus 209 controls the flow of data in the information processing apparatus 106.

  In the present embodiment, for convenience of explanation, a configuration in which the information processing apparatus 106 is realized by one apparatus will be described, but it may be realized by a configuration in which resources are distributed to a plurality of apparatuses. For example, storage and calculation resources may be distributed in a plurality of devices. Alternatively, resources may be distributed for each component virtually realized on the information processing apparatus 106 to perform parallel processing.

  Further, the information processing apparatus 106 does not need to include all the components described here. For example, when it is not necessary to input data from the media 204, the external storage drive 203 can be omitted.

<Hardware configuration of POS terminal 110>
FIG. 3 is a functional block diagram showing a hardware configuration of the POS terminal 110 used in the present embodiment.

  The CPU 300 controls the POS terminal 110 as a whole by executing a control program (firmware) or the like stored in the SRAM 303.

  A ROM 301 is a non-volatile memory that stores programs such as a basic I / O program and various data such as font data used for receipt printing.

  A RAM 302 is a volatile memory for temporarily storing various data, and functions as a main memory, a work area, and the like of the CPU 300.

  The SRAM 303 is a non-volatile memory that stores a control program for the POS terminal 110, sales data, and the like. The SRAM 303 also stores the sampling condition instructed from the information processing apparatus 106, and the control program determines whether to print the questionnaire request on the receipt according to this condition. Note that the above programs, data, and the like may be stored using an HDD or the like instead of the SRAM 303.

  The printing unit 304 is for printing a receipt.

  The input unit 305 is used to input data to the POS terminal 110, and includes a barcode reader, a keyboard, and the like.

  The display 306 is for displaying the contents of data input from the input unit 305 and the state of the POS 110 (such as “standby”).

  An interface (I / F) 307 is configured by a network I / F or the like, and is connected to the Internet 100 via this to exchange data with the information processing apparatus 106 or the like.

  The system bus 308 manages the flow of data in the POS terminal 110.

  Note that the POS terminal 110 may be configured in such a manner that resources are distributed to a plurality of devices, as in the case of the information processing device 106.

<Overview of how to collect questionnaire responses>
With reference to FIG.4 and FIG.5, the outline | summary of the process which the information processing apparatus 106 collects a questionnaire reply result from a customer is demonstrated.

  FIG. 4 is a diagram illustrating an example of a receipt 400 printed by the printing unit 304 of the POS terminal 110. FIG. 4A shows the surface of the receipt 400, which is the same as a general receipt. FIG. 4B is the back side of the receipt 400, and the questionnaire request is described here.

  When the customer makes a purchase at the survey store, the POS terminal 110 determines whether to print a questionnaire request on the back side of the receipt 400 according to the sampling conditions. As sampling conditions, for example, “every 10 people”, “when 500 yen or more is purchased”, “when a predetermined product is purchased”, and the like can be considered. Of course, the questionnaire request may be printed on the receipts 400 of all customers.

  Therefore, the questionnaire request shown in FIG. 4B is printed on the back side of the receipt 400 of the customer that satisfies the sampling condition, and the back side of the receipt is blank (or other advertisement) for other customers.

  The predetermined condition is stored in the SRAM 303, but can be changed by the orderer via the PC 108 and the information processing apparatus 106.

  As shown in FIG. 4B, at least one of URL 401, telephone number 402, and two-dimensional barcode 403 is described in the questionnaire request. Based on such information, the customer can answer the questionnaire via the mobile terminal 102 or the PC 104.

  A receipt number 404 is also written on the back side of the receipt 400. The receipt number 404 is associated with information (hereinafter referred to as “receipt information”) such as a product purchased by the customer, an amount of money, a store visit time, a store visit day, and the like. When the POS 110 issues a receipt 400 with a questionnaire request, the receipt information is transmitted to the information processing apparatus 106 in association with the receipt number 404, and the information processing apparatus 106 records it in the HDD 205. In the present embodiment, the description will be made on the assumption that there is one survey target store. However, in practice, a plurality of stores can be combined into a survey target store.

  FIG. 5 is a diagram illustrating an example of an answer screen 500 displayed on the display of the PC 104 when the customer accesses the URL 401 using the PC 104.

  The answer screen 500 includes questions (501 to 506 and 510) in an option (choice) format, a question (507) in a free description format, a send button 511, and a cancel button 512. Quantitative answers can be obtained from choice-type questions, and qualitative answers can be obtained from free-form questions.

  The customer answers the question in the option format by checking a radio button or selecting one candidate from a pull-down menu. The questions 501 and 510 are for the customer to input arbitrary values. However, since the range is limited, the questions 501 and 510 are handled in the same row as the question in the option format from which a quantitative answer is obtained. Customers also answer free-form questions by text entry.

  When the customer clicks the send button 511 using the mouse of the PC 104 or the like, questionnaire response information (hereinafter simply referred to as “response information”) is transmitted to the information processing apparatus 106.

  Here, the information processing apparatus 106 can know the information of the questions 502 and 503 from the receipt information associated with the receipt number 404 of the question 501 regardless of the customer input. Therefore, the information on the questions 502 and 503 is used for the purpose of detecting illegal answers. That is, the application program of the information processing apparatus 106 collates the receipt information with the response information, and if the degree of contradiction exceeds a predetermined threshold value, treats the response as an illegal response and does not use it for the following analysis.

  When the customer calls the telephone number 402, for example, the operator verbally asks the same question as the question on the answer screen 500 and inputs the answer on a PC (not shown). By transmitting this answer to the information processing apparatus 106, the information processing apparatus 106 can collect the answer information. Of course, the operator may input the answer information directly into the information processing apparatus 106 using the instruction input apparatus 206 of the information processing apparatus 106.

  The two-dimensional barcode 403 includes at least information indicating the URL 401 or the telephone number 402.

  As described above, the information processing apparatus 106 acquires qualitative information and quantitative information as response information from the mobile terminal 102 or the PC 104. Further, the information processing apparatus 106 can acquire qualitative information and quantitative information from the POS terminal 110 by using the receipt number 404 acquired from the mobile terminal 102 or the PC 104.

<Generation of element table>
FIG. 6 is a diagram showing an element table 600 generated by the information processing apparatus 106 based on the answer information. The element table 600 is used when the information processing apparatus 106 generates a report based on the analysis of the Internet survey. The element table 600 generated by the information processing apparatus 106 is recorded in the HDD 205, for example.

  The sample number 601 is an identification number assigned to each respondent. The sample number 601 may be a serial number as long as it can identify the respondent. For example, the receipt number 404 may be used as the sample number 601 as it is.

  Elements 602 to 617 correspond to the answers to the questions 502 to 510 in FIG. Further, although not shown, the information acquired by the information processing apparatus 106 from the POS terminal 110 can also be included in the element table 600. That is, the “element” includes a question that is directly asked to the respondent using the answer screen 500 and a question that is indirectly asked to the respondent via the POS terminal 110 (for example, a question “What is the purchased product?”). It is a concept that includes. Accordingly, for example, the element 603 includes “Saturday (day)”, “Sunday (day)”, and the like as specific answers. Strictly speaking, these should be referred to as “answer information included in the element 603”. However, in the following, these answer information are also referred to as elements, and may not be strictly distinguished.

  Here, in the elements 605 to 613, 1 indicates “very satisfied”, 2 indicates “somewhat satisfied”, 3 indicates “somewhat dissatisfied”, and 4 indicates “very dissatisfied”. In the element 615, 1 indicates “once a week”, 2 indicates “a few times a month”, 3 indicates “once a month”, 4 indicates “less than that”, and 5 indicates “this is the first time”.

  The reason why the question 501 (receipt number 404) is not included is that the receipt number 404 is used for detecting an illegal answer as described above and is not used for analyzing the answer information. Of course, for example, when the receipt number 404 is used for analysis by using the fact that the smaller the receipt number 404 is, the earlier the customer visits the store, the receipt may be included in the element table 600.

  Further, the element table 600 is not limited to the information obtained from the questionnaire in FIG. 5, and may include receipt information (for example, purchased products) that can be understood from the receipt number 404.

  The element 614 describes only specific words / phrases among the free answers to the question 507, but actually includes all the words that share the free answers. For example, for the sample 001, the element 614 includes nine words of “car”, “no”, “handling”, “ga”, “careful”, “de”, “work”, “ga”, and “speedy”. That is, the information processing apparatus 106 can decompose the free answer into words by using, for example, automatic segmentation.

  Automatic segmentation refers to dividing a target processing unit into sentences in a morphological process of a natural language sentence by a computer. Character string matching between the heading of the word dictionary and the input sentence is the basis of processing, and refraction processing and derivation processing are used. The Japanese splitting procedure includes the longest matching method and the minimum number of clauses that divides the sentence so that the number of phrases in the sentence is minimized (Masao Nagao, Haruhisa Ishida and others, “Iwanami Information Science Dictionary”, 1990, Iwanami Shoten).

  An example of software having the function of automatic division writing is Happiness / AiBASE of Peace Information Center Co., Ltd., for example.

  Then, when actually performing the analysis work, use the user dictionary (in this example, for the CS survey of the gas station) to select “Polite” and “Speedy” and use only these words. Is determined.

  Although details will be described later, it is not necessary to use all the elements of the element table 600 when analyzing the answer information. For example, if the orderer is not interested in the customer's purpose of use, the element 604 may be excluded from the analysis target.

  As described above, there may be a plurality of investigation target stores. In this case, the element table 600 may include a store number, store characteristics (region, scale, location conditions, etc.), and the like.

<Order report>
FIG. 7 is a diagram showing an order form 700 used when an orderer orders a report based on the Internet survey. The order form 700 may be displayed on the display of the PC 108 and transmitted to the information processing apparatus 106 via the Internet 100, or may be transmitted to the operator of the information processing apparatus 106 via FAX or the like. In the case of using FAX, the operator of the information processing apparatus 106 inputs items described in the order form 700 to the information processing apparatus 106 by using the instruction input device 206 or the like.

  An item 701 is for selecting a report type required by the orderer. Here, types 702 to 706 are defined in advance. For example, when the type 702 is selected, the information processing apparatus 106 generates a report representing the degree of satisfaction for each sex with respect to the service and provides the report to the orderer. In type 702, only quantitative information is used, and qualitative information is not used. Therefore, it is irrelevant to the object of the present invention, “easy and integrative handling of qualitative information and quantitative information”, and detailed description thereof is omitted. The same applies to the types 703 and 704.

  When the type 705 is selected, the information processing apparatus 106 uses the free answer (element 614) to indicate, for example, how many positive opinions and how many negative opinions exist regarding the speed of the service. Generate a report. Qualitative information is used in type 705, but quantitative information is not used. Therefore, as in types 702 to 704, this is irrelevant to the object of the present invention, and detailed description thereof is omitted.

  When the type 706 is selected, the information processing apparatus 106 uses the element corresponding to at least one qualitative question and the element corresponding to at least one quantitative question among the elements of the element table 600 to report. Generate. Further, the information processing apparatus 106 uses not only a question directly displayed on the answer screen 500 but also an element corresponding to a question (for example, “What is the purchased product?”) That is indirectly asked about the POS terminal 110. Also good. Thereby, combined use (hybridization) of qualitative information and quantitative information is performed. Details of the report generation will be described later.

  Although five types of reports of types 702 to 706 are defined here, any number may be used in practice. However, it is preferable that there is only one type of report defined in type 706.

  The item 707 is used when the orderer needs a report based on an element other than the type specified in the item 701 (breakdown, BD). Since the item 707 is also irrelevant to the object of the present invention, detailed description is omitted.

  Item 708 is used to select an element to be used for analysis when type 706 is selected in item 701. The elements used for the analysis are selected from the element table 600 and include at least one qualitative information (element) and at least one quantitative information (element). In FIG. 7, although there are eight entry columns for the used elements, any number can be used. In practice, it is preferable that all the elements in the element table 600 have a selectable number. Here, when the element 614 which is qualitative information is selected, as described above, which word is used is selected using the user dictionary.

  Item 709 is used to indicate the delivery date of the report.

  The item 710 is used to notify the operator of the information processing apparatus 106 of other demands of the orderer, such as the number of response information (number of respondents) required by the orderer.

<Clustering respondents>
Next, the manner in which answer information is analyzed based on the element table 600 and the order form 700 and respondents are clustered will be described.

  For the analysis, a corresponding analysis method well known in the field of statistics is used. Briefly, first, all the elements selected in the item 708 are arranged in a multidimensional space. The number of dimensions is sufficient to classify respondents based on experience.

  FIG. 8 is a diagram illustrating a cluster arrangement 800 that conceptually represents clusters. As shown in the figure, in this embodiment, the respondent is classified into six clusters 811 to 816. The cluster layout 800 is generally three or more dimensions in actuality, but is two-dimensional here in order to simplify the drawing. In the example shown in FIG. 8, there are elements related to “appearance” such as “new building” and “clean interior” in the upper part of axis 1 (851), and “easy access” and “comfort of rest room” in the lower part. There is an element related to the “use feeling” of a specific facility, such as “Good”. Axis 2 (852) The upper part, that is, the right side, has elements related to “price factors” such as “Easy to display price” and “low price”, and the lower part, that is, the left side, “Correspondence by salespeople”, “Service is speedy” There are elements related to “human service evaluation”. In other words, it is possible to verify that the tendency (direction) of the elements to be evaluated is different depending on the respondent, or that a cluster showing the same answer pattern exists. Respondent clustering means that respondents with similar characteristics (similar response patterns) to multi-dimensional component scores (axis) are “statistically classified” into several groups (clusters) ( The distance between the clusters is significant). All respondents are classified into any specific cluster, there is no respondent who does not belong to any cluster, and there is no respondent who belongs to two or more clusters. In addition to newly discovered clusters, if you specify attributes (ie clusters) such as “female / 30's” or “total satisfaction / high” by gender, you can also observe the “respondent group” belonging to them. It becomes.

  The clustering of respondents is performed only when a hybrid report type (type 706 in the example of FIG. 7) is selected in the item 701 of the order form 700.

  The elements specified in the item 708 of the order form 700 are used for clustering the respondents. The elements 821 to 832 represent a part of the designated elements. Although not shown, all the designated elements are actually arranged at any location on the cluster arrangement diagram 800. As described above, the elements 821 to 832 are strictly “answer information included in the elements”.

  Further, although only the elements based on the qualitative information are shown in the cluster layout diagram 800, actually, elements based on quantitative information such as the element 606 in FIG. 6 which is an answer to the question 505b in FIG. 5 are also arranged. . The answers to the question 505b may include 1 to 4, that is, “very satisfied” to “very dissatisfied”, but each is arranged at a predetermined position.

  Each cluster has characteristic information generated based on a reference (axis) on which the cluster is generated. For example, respondents classified into the cluster 811 generally have a positive opinion about the newness and cleanliness of the building. Therefore, the cluster 811 uses the characteristic information “satisfied with the cleanliness of the building” as characteristic information. Have.

  Further, each cluster has attribute information that is generated based on the contents of elements that appear prominently for the respondents classified into the cluster. For example, it is assumed that most of the respondents classified in the cluster 811 are in their 30s (element 617) and female (element 616). In this case, a male other than a woman in their 30s, for example, a man in their 40s may belong to the cluster 811. However, the cluster 811 has attribute information “female in its 30s” as indicating a general tendency. Note that the elements used for generating the attribute information may be those used for clustering or not used.

  The attribute information is generated by determining whether the center of gravity of each element (for example, element 616 “gender”) held by the respondent classified into the target cluster is close to a specific content (for example, “female”). Is done. If the center of gravity of the element 616 “gender” is close to “female”, the content of the element that appears prominently for the respondents classified in the cluster is “female”. The same process is performed for all elements used for generating attribute information, and attribute information such as “30s, women” is finally obtained.

  It should be noted that not all clusters necessarily have elements that appear prominently.

  In the example of FIG. 8, the number of clusters is six. However, in actuality, the number of stores to be surveyed, the number of respondents, etc. in the survey field (for example, a CS survey at a gas station or a CS survey at a convenience store) Accordingly, numbers that are empirically recognized as being suitable for obtaining useful analysis reports are used. However, the information processing apparatus 106 may be configured so that the number of clusters can be explicitly specified in the order form 700.

<Presentation of analysis results>
FIG. 10 is a diagram showing an analysis report 1000 generated based on information generated based on the element table 600 and the order form 700 (that is, clusters 811 to 816). The analysis report 1000 is presented to the orderer, who can view it using the PC 108. Of course, the analysis report 1000 may be presented to the orderer by being printed on paper.

  The analysis report 1000 includes at least two fields, a characteristic information field 1001 and a ratio field 1002. Optionally, an attribute information field 1003 and an advice field 1004 are included.

  The characteristic information field 1001 is a field indicating characteristic information generated based on the elements included in each cluster, as described with reference to FIG.

  A ratio field 1002 is a field indicating the ratio of respondents belonging to each cluster. As described above, since all the respondents are always classified into one cluster, the total ratio in the ratio field is 100%. Of course, it may not be divisible when calculating the ratio, and may be slightly different from 100% as a result of rounding off.

  As described with reference to FIG. 8, the attribute information field 1003 is a field indicating attribute information generated based on the contents of elements that appear prominently among respondents classified into each cluster. As described above, for example, respondents who belong to the cluster 811 remarkably appear as women in their 30s. That is, the attribute centroid is close to the cluster centroid. Further, like the cluster 814, there may be a cluster in which there is no noticeable element and the attribute information is blank.

  The advice field 1004 is a field indicating advice to the orderer. The advice is generated by using the characteristic information of each cluster and optionally the ratio and attribute information, and indicates a countermeasure plan that the orderer should take in order to improve CS. For example, in the cluster 812, men in their early twenties are dissatisfied with the service speed, and it is considered that the service speed needs to be improved particularly during the early morning hours. Therefore, the corresponding advice is “service should be done particularly quickly in the early morning”. In the analysis report 1000, advice is given only to clusters that are dissatisfied with something, but in order to further improve CS, information is given so that advice is given to clusters that are satisfied with something. The processing device 106 may be configured.

  An example of a method by which the information processing apparatus 106 generates advice will be described in more detail. The HDD 205 of the information processing apparatus 106 holds an advice table including a large number of advices associated with characteristic information, attribute information, and the like. FIG. 9 is a diagram illustrating an example of the advice table 900. The information processing apparatus 106 examines characteristic information and the like for each cluster, and selects a corresponding advice from the advice table. For example, since the cluster 815 has the characteristic information “dissatisfied with the service” and the attribute information “female”, the advice of the entry 903 “need to improve the service for women as a whole” is selected. The

<Process flow>
FIG. 11 is a flowchart showing a flow of processing performed by the information processing apparatus 106 in the present embodiment. The processing in each step of the flowchart is performed by the CPU 200 in FIG. 2 executing an application program stored in the HDD 205.

  In step S <b> 1101, the information processing apparatus 106 acquires order information from the PC 108 and stores it in the HDD 205. More specifically, when the information processing apparatus 106 receives an Internet survey request from the PC 108 via the Internet 100, the information processing apparatus 106 transmits an ordering form 700 to the PC 108. The orderer fills in an order form 700 on the PC 108 and transmits it to the information processing apparatus 106. In the present embodiment, it is assumed that type 706 is selected in item 701 of ordering form 700.

  Note that the processing in step S1101 may be performed by FAX, mail, or the like without using the Internet 100. In this case, the operator of the information processing device 106 uses the instruction input device 206 to input order information to the information processing device 106.

  In step S1102, the information processing apparatus 106 notifies the sampling condition to the POS terminal 110 of the investigation store (in this embodiment, a gas station), and sets it in the SRAM 303. Sampling conditions may be configured so as to be stored in the HDD 205 in advance according to the survey field, or may be designated by the orderer using the item 710 of the order form 700.

  The POS terminal 110 is generally owned by the orderer and may not be under the jurisdiction of the operator of the information processing terminal 106. In this case, instead of the information processing terminal 106, the PC 108 may notify the POS terminal 110 of the sampling condition and set it in the SRAM 303.

  In step S <b> 1103, the information processing apparatus 106 collects response information from the mobile terminal 102 or the PC 104. More specifically, when receiving an access request using the URL 401 or the like, the information processing apparatus 106 transmits an answer screen 500 to the mobile terminal 102 or the PC 104. The respondent uses the portable terminal 102 or the PC 104 to transmit answer information for each question on the answer screen 500 to the information processing apparatus 106. At the same time, receipt information may be acquired from the POS terminal 110. In the following, unless otherwise specified, the “response information” may include “receipt information”.

  Even when there is a margin until the delivery date specified in the item 709 of the order form 700 or when the required number of respondents instructed in the item 710 is less, the information processing apparatus 106 collects a predetermined number of questionnaire response information. The process proceeds to step S1104. Although the details of this reason will be described later, in brief, it is for enabling real-time analysis of answer information. Therefore, if real-time analysis is not required, the process may proceed to step S1104 after collecting all answer information required by the information processing apparatus 106. The “predetermined number” is preferably the minimum number necessary for the analysis described below.

  When the respondent intends to answer the questionnaire using the telephone number 402, for example, the operator who receives the call transmits the reply content to the information processing apparatus 106 using a PC (not shown).

  In step S1104, the information processing apparatus 106 generates an element table 600 based on the answer information collected in step S1103. As described above, at this stage, the element 614 includes all the words in which the answer sentence to the question 507 is automatically segmented.

  In step S1105, the information processing apparatus 106 clusters respondents based on the element table 600. The elements that are actually used for clustering are specified in the item 708 of the order form 700 and have already been acquired in step S1101. In addition, the information processing apparatus 106 uses an unnecessary word among the words included in the element 614 by using the user dictionary stored in the HDD 205 corresponding to the survey field (here, the CS survey of the gas station). Is excluded. Furthermore, as described above, the information processing apparatus 106 performs clustering of respondents using the correspondence analysis method.

  In step S1106, the information processing apparatus 106 generates an analysis report 1000 based on the cluster generated in step S1105. If the analysis report 1000 has already been generated, the information processing apparatus 106 updates the content of the analysis report 1000. However, when new questionnaire response information is not collected in step S1103, the information processing apparatus 106 does not update the content of the analysis report 1000 because there is no change in the cluster generated in step S1105.

  In step S1107, the information processing apparatus 106 presents the analysis report 1000 to the orderer. Specifically, when the information processing apparatus 106 receives a presentation request for the analysis report 1000 from the PC 108 via the Internet, the information processing apparatus 106 transmits the analysis report 1000 to the PC 108. Alternatively, the information processing apparatus 106 may be configured to periodically transmit the analysis report 1000 to the PC 108 by e-mail or the like without receiving a presentation request.

  Note that the processing in step S1107 may be performed by FAX, mail, or the like without using the Internet 100, similarly to the processing in step S1101.

  In step S1108, the information processing apparatus 106 determines whether the Internet survey satisfies a predetermined end condition. The predetermined condition is, for example, a condition specified on the ordering form 700 regarding the survey period, the number of questionnaire response information to be collected, and the like. If the conditions are met, the Internet survey is terminated. If the condition is not satisfied, the process returns to step S1103, and the information processing apparatus 106 further collects questionnaires, and again analyzes and presents the analysis report 1000 by the processes of steps S1104 to S1107. Thereby, the analysis report 1000 is updated in real time or at a predetermined cycle, and the orderer can obtain the latest and more useful analysis report.

<Sequential presentation of analysis reports>
With reference to FIG. 12, the flow of processing for sequentially (rapidly) notifying the orderer that the analysis report 1000 has been updated will be described.

  In step S1201, the information processing apparatus 106 determines whether a predetermined period has elapsed. The predetermined period is, for example, a period such as “one day”, “one week”, “one month”, and can be designated by the orderer using the item 710 of the order form 700 or the like. If the predetermined period has elapsed, the process proceeds to step S1202.

  In step S1202, the information processing apparatus 106 determines whether or not the analysis report 1000 has been updated. The analysis report 1000 is updated in step 1106 of FIG. 11 as described above. If the analysis report 1000 has been updated, the process proceeds to step S1203.

  In step S1203, the information processing apparatus 106 transmits to the PC 108 a notification that the analysis report 1000 has been updated. This notification is transmitted via the Internet 100 using e-mail, for example. Thus, the orderer can quickly know that the analysis report 1000 has been updated.

  When it is determined in step S1202 that the analysis report 1000 has not been updated, the information processing apparatus 106 may transmit a notification to the PC 108 that the analysis report 1000 has not been updated. As a result, regardless of whether or not the analysis report 1000 is updated, the orderer can know the state of the latest analysis report 1000 every predetermined period.

<Summary of Embodiment>
As described above, according to the present embodiment, the information processing apparatus 106 generates the element table 600 from the qualitative information and quantitative information included in the answer information obtained from the questionnaire. In addition, the information processing apparatus 106 clusters respondents using the element specified by the item 708 of the order form 700. Next, the information processing apparatus 106 provides the orderer with an analysis report 1000 that includes information on the characteristic information and ratio of each cluster.

  As a result, the information processing apparatus 106 can easily and integratively handle qualitative information and quantitative information, and can generate and provide a more reliable analysis report of an Internet survey.

It is a figure which shows the outline | summary of the structure of the system used in an internet investigation. 2 is a functional block diagram showing a hardware configuration of an information processing apparatus 106. FIG. 2 is a functional block diagram showing a hardware configuration of a POS terminal 110. FIG. 6 is a diagram illustrating an example of a receipt 400 printed by a printing unit 304 of a POS terminal 110. FIG. It is a figure which shows an example of the reply screen 500 displayed on the display of PC104 when a customer accesses URL401 using PC104. It is a figure which shows the element table 600 which the information processing apparatus 106 produces | generates based on reply information. It is a figure which shows the order form 700 used when an orderer orders the report based on an internet investigation. FIG. 7 is a diagram illustrating a cluster configuration diagram 800 that conceptually represents a cluster. 5 is a diagram illustrating an example of an advice table 900. FIG. It is a figure which shows the analysis report 1000 produced | generated based on the information (namely, cluster 811-816) produced | generated based on the element table 600 and the order form 700. FIG. 6 is a flowchart showing a flow of processing performed by the information processing apparatus 106. It is a flowchart which shows the flow of a process which notifies a purchaser sequentially (rapidly) that the analysis report 1000 was updated.

Claims (11)

An information processing apparatus connected to a plurality of client terminals via the Internet,
Obtaining first qualitative information and first quantitative information as answer information for qualitative questions and quantitative questions from a plurality of respondents who are users of the plurality of client terminals via the plurality of client terminals Acquisition means to
Decomposition means for decomposing the first qualitative information into words and generating second qualitative information;
First selection means for selecting a predetermined question from the qualitative question and the quantitative question;
A second selection means for selecting a predetermined word as the third qualitative information among the second qualitative information corresponding to the selected qualitative question;
Clustering means for classifying the plurality of respondents into a plurality of clusters based on the second quantitative information corresponding to the quantitative question selected from the quantitative information and the third qualitative information;
An information processing apparatus comprising:   The clustering means classifies the respondents into the plurality of clusters based on the similarity between the information included in the second quantitative information and the third qualitative information for each respondent. The information processing apparatus according to claim 1.   The information processing apparatus according to claim 1, wherein the clustering unit performs the classification using a correspondence classification method.   Furthermore, based on the first quantitative information and the second qualitative information of the respondent classified into each of the plurality of clusters, the tendency of the respondent classified into the cluster for each of the plurality of clusters The information processing apparatus according to claim 1, further comprising attribute information generation means for generating attribute information indicating A report generating means for generating an analysis report from a ratio of the respondents classified into each of the plurality of clusters to a total of the plurality of respondents;
The information processing apparatus according to claim 4, wherein the attribute information is included in the analysis report. Further, based on the second quantitative information and the third quantitative information, which are the criteria for classification by the clustering means, and the attribute information, advice information on how to deal with the respondent included in the cluster Advice information generating means for generating
The information processing apparatus according to claim 5, wherein the advice is included in the analysis report.   The information processing apparatus according to claim 5, further comprising notification means for notifying an external device at a predetermined interval that the analysis report has been generated or updated by the report generation means. The information processing apparatus is further connected to a second client terminal other than the plurality of client terminals via the Internet,
The acquisition unit acquires at least one of the first qualitative information and the first quantitative information also from the second client terminal. Information processing device. A method for controlling an information processing apparatus connected to a plurality of client terminals via the Internet,
Obtaining first qualitative information and first quantitative information as answer information for qualitative questions and quantitative questions from a plurality of respondents who are users of the plurality of client terminals via the plurality of client terminals An acquisition process to
A decomposing step of decomposing the first qualitative information into words to generate second qualitative information;
A first selection step of selecting a predetermined question from the qualitative question and the quantitative question;
A second selection step of selecting a predetermined word as third qualitative information from among the second qualitative information corresponding to the selected qualitative question;
A clustering step of classifying the plurality of respondents into a plurality of clusters based on second quantitative information corresponding to the quantitative question selected from the quantitative information and the third qualitative information;
A control method comprising: A program for causing a computer to execute a control method for an information processing apparatus connected to a plurality of client terminals via the Internet, the control method comprising:
Obtaining first qualitative information and first quantitative information as answer information for qualitative questions and quantitative questions from a plurality of respondents who are users of the plurality of client terminals via the plurality of client terminals An acquisition process to
A decomposing step of decomposing the first qualitative information into words to generate second qualitative information;
A first selection step of selecting a predetermined question from the qualitative question and the quantitative question;
A second selection step of selecting a predetermined word as third qualitative information from among the second qualitative information corresponding to the selected qualitative question;
A clustering step of classifying the plurality of respondents into a plurality of clusters based on second quantitative information corresponding to the quantitative question selected from the quantitative information and the third qualitative information;
A program comprising:   The computer-readable storage medium which recorded the program of Claim 10.

Images