JP4264809B2 - Free opinion analysis system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a system for processing a free description sentence (hereinafter referred to as a free opinion), for example, extracting a customer need from a free opinion obtained by a questionnaire survey or the like, and further using it to come up with an idea. Related to the free opinion analysis system.
[0002]
[Prior art]
As a conventional free opinion analysis system, there is a document processing apparatus that analyzes a questionnaire response as a result of a questionnaire when the response is made in a natural sentence (see, for example, Patent Document 1).
[0003]
[Patent Document 1]
JP 2002-140465 A (page 3-8, FIG. 1)
[0004]
[Problems to be solved by the invention]
The conventional document processing device uses the words that are simply used, that is, aggregates and displays the corresponding words by keyword matching, etc. Judgment processing is performed for each meaning, and totalization and display are performed for each evaluation. However, at the current technical level of text mining, the classification accuracy of mechanical automatic analysis is lower than the classification accuracy by humans (flexibility for ambiguous concepts), and mechanical automatic analysis should be classified with the same level of accuracy as humans. There is a problem that classification results cannot be used as they are. For example, in a questionnaire on air conditioners, it is difficult to categorize “the feet are cold” and “the heating feet were dissatisfied with the feet” into the same category by automatic mechanical analysis. The results obtained by the above are word level tabulation results, for example, “Footsteps: 98” and “Cold: 120”, so only the superficial results are known and background information useful for future product development is obtained. Is difficult.
[0005]
In addition, humans have an advanced information processing function that recalls ideas and ideas by stimulating information from the outside world through the perception-recognition-motion process and organicizing it. As a result, the entire system can obtain an output that exceeds the input data. However, the conventional document processing apparatus only uses the original input data, and the results obtained are within the range. There is a problem that it is not supported so as to support the idea generation by humans based on the result of mechanical automatic analysis.
[0006]
Furthermore, conventional document processing devices do not have a configuration that takes into account collaborative work by multiple workers. For example, when developing new products based on free opinions, free opinions that are the voice of customers Cannot share the development policy vector at the same time among the parties involved in product development, and utilize the various knowledge of parties involved in development (eg design, product certification, sales, etc.) From the viewpoint of workability (speed), it is impossible to analyze customer needs and recall ideas for product development and sales promotion, and the current product development cycle is becoming shorter. There is a problem that efficient processing by simultaneous parallel work cannot be performed.
[0007]
The present invention was made to solve the above-described problems, and can edit a mechanical automatic analysis result using human knowledge to perform a more accurate free opinion analysis. In addition, ideas can be supported for mechanical automatic analysis results and human editing results. In addition, customer opinions can be shared among multiple analysts, and ideas can be effectively and efficiently developed through collaborative work. To obtain a free opinion analysis system that can be implemented.
[0008]
[Means for Solving the Problems]
In the free opinion analysis system according to the present invention, a data input unit that inputs at least data from a free description sentence, a data table unit that stores data input from the data input unit, and the data input unit A data processing unit for classifying the collected data, an editing unit having a function of re-editing a result obtained by the data processing unit, and a mark or a mark for the processing result by the data processing unit or the editing unit An idea idea support unit having at least a comment input function of comment or character editing , a reporting unit that processes the contents of each part into a report format, and an output unit that displays the contents of each part , the idea idea support The section has a comment title database in which a plurality of comment titles are set, To sense a result, the comment that the title selected by the operation input from the comment title database, add as an attribute for classifying the data, a comment input, and inputs as attribute data of the attribute.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Embodiment 1 FIG.
FIG. 1 is an overall configuration diagram of the free opinion analysis system showing Embodiment 1 of the present invention, FIG. 2 is a diagram showing a specific example of data stored in the data table of this free opinion analysis system, and FIG. 3 is this free opinion FIG. 4 is a diagram schematically showing the contents of the data table of the analysis system, FIG. 4 is a diagram showing the classification result of data of this free opinion analysis system, and FIG. 5 is a data table of the configuration and processing results of the data processing unit of this free opinion analysis system FIG. 6 is a diagram showing the configuration of the data mining unit of this free opinion analysis system, and FIG. 7 is a diagram showing the processing results of the data mining unit of this free opinion analysis system.
[0010]
FIG. 8 is a diagram showing another processing result of the data mining unit of the free opinion analysis system, FIG. 9 is a diagram showing a configuration of the editing unit of the free opinion analysis system, and FIG. 10 is an editing work of the free opinion analysis system. FIG. 11 is a diagram showing the update of the data table, FIG. 11 is a diagram showing the editing screen of this free opinion analysis system, FIG. 12 is a diagram showing the configuration of the idea idea support unit of this free opinion analysis system, and FIG. 13 is this free opinion analysis system FIG. 14 is a diagram showing the result of editing by the idea idea support unit of FIG. 14, FIG. 14 is a diagram showing the update of the data table by the idea idea support unit of this free opinion analysis system, and FIG. 15 is a diagram showing the configuration of the reporting unit of this free opinion analysis system 16 to 19 are diagrams showing examples of output from the reporting unit of this free opinion analysis system.
[0011]
In the figure, 1 is a data input unit consisting of an input unit for inputting a physical means of a keyboard for directly inputting a free opinion obtained by a questionnaire survey or the like, a file obtained by a web survey, etc., and 2 is input Data table for storing data, 3 is a data processing unit that searches and classifies stored free opinion data according to their meaning, etc. 4 is data mining that analyzes various data, such as arithmetic calculation, statistical analysis, multivariate analysis, etc. Sections 5 and 5 are editing sections for humans to correct and edit the processing results obtained by the data processing section 3 and the data mining section 4.
[0012]
6 is an idea idea support unit for assisting an analyst to come up with an idea using the result of processing or the result of correction / editing by an analyst, 7 is a data processing result, the result of correction / editing by an analyst, an idea idea A reporting unit for processing the results into a report format, 8 a display for displaying the contents of the data table 2, the data processing unit 3, the data mining unit 4, the editing unit 5, the idea generation support unit 6, and the reporting unit 7 And so on.
[0013]
Next, the operation will be described.
First, data or a file is input by the data input unit 1 and stored in the data table 2. A specific example of data stored in the data table 2 is shown in FIG. In addition to the free opinion, the input data here can be data such as option category data, numeric data such as age, ranking data representing ranking, multi-answer data with multiple selections, graphical data such as pictures and photos, etc. All data formats are targeted. In this way, individual free opinions are numbered and stored.
[0014]
FIG. 3 schematically shows the contents of the data table 2. The input data is stored as “numbering”, “free opinion”, and data other than the free opinion as “attribute data” for each record (one horizontal line).
When data is stored in the data table 2, the data processing unit 3 classifies the data based on the content of the free opinion. FIG. 4 is an example in which the classification results are displayed hierarchically. In the case of hierarchization, the hierarchy number is added to the data table 2 as shown in FIG. In this example, the free opinion: “cool well” corresponds to the hierarchy number “112”, which means that the first hierarchy “1” (cooling performance) and the second hierarchy “1” (cooling ○). It is included and is located in the third hierarchy “2”.
[0015]
As shown in FIG. 6, the data mining unit 4 includes arithmetic calculation, statistical analysis, multivariate analysis, and various other data analysis functions, and an output format creation unit for displaying it on a table, graph, diagram, etc. For example, FIG. 7 shows the appearance frequency of the words used in the free opinion by the data mining unit 4 and the appearance frequency thereof is calculated, and the result is displayed in a ranking by a horizontal bar graph. FIG. 8 shows a certain word, here In the ranking display of FIG. 7, we focused on “outdoor units” that appeared relatively frequently, and ranked words used by frequency in sentences containing the word “outdoor unit” by frequency (cross analysis). It is. Here, by displaying the words used at the same time, it is possible to roughly grasp how the outdoor unit is evaluated.
As described above, the data stored in the data table 2 is searched / classified by the data processing unit 3 and aggregated by the data mining unit 4. Depending on the purpose of the analysis, various data such as a table format, a graph format, and a diagram format are available. Can be displayed in any form.
Note that the analysis results obtained so far are results automatically processed by the data processing unit 3 and the data mining unit 4 to the last.
[0016]
Therefore, in this system, in order to provide a system that enables analysis of free opinions with higher accuracy using the knowledge of the analyst, the editing unit 5, the automatic mechanical analysis result, and the editing result by the analyst, In order to be able to support the idea, an idea idea support unit 6 is provided.
First, the configuration of the editing unit 5 will be described.
As shown in FIG. 9, the editing unit 5 re-edits the mechanical classification result by drag and drop (upper parallel, lower parallel, parent hierarchy, child hierarchy), copy and paste (copy, cut, paste, insert). , Delete) functions, and freely move and edit between hierarchical structures into which free opinions are classified, and can categorize and divide opinions according to the KJ method. This operation can also be performed using a menu selection and a shortcut using the control key.
[0017]
Next, the operation of the editing unit 5 will be described.
First, FIG. 11 is an editing screen before editing. Here, “cooling capacity and dehumidification capacity” has two meanings of “cooling” and “dehumidification”, and the analyst has two opinions. The “dehumidification ability preeminent” is moved to the category of “dehumidification” and “dehumidification ○”. As a result of this editing operation, the hierarchy number of the contents of the data table 2 is updated as shown in FIG. Further, when the free opinion itself is changed, an attribute “free opinion new” is newly added.
In this way, the editing unit 5 can record the newly modified free opinion content while retaining the original free opinion content, and can also retain the hierarchical structure thereof.
[0018]
Next, the idea generation support unit 6 will be described.
The analyst looks at the results processed by the processing unit 3 and the data mining unit 4, discovers an opinion of interest through KJ-based editing work by the editing unit 5, and recalls a new idea. In order to facilitate this, the support unit 6 uses a symbol Data Base (hereinafter referred to as “D. Base”) to mark the opinions of interest (★, ☆, ○, ×, Δ, etc.) as shown in FIG. B.), a comment title D. for entering the recalled idea. B. Furthermore, it has a character editing function for setting the font, size, color, style, and the like. These input / setting operations can also be performed by shortcuts using menu selection and control keys.
[0019]
Next, as an example of the editing result of the idea idea support unit 6, FIG. B. A mark added in accordance with B. Comment title B. Select “keyword” from B and enter a comment. B. Shows that "request" and "Idea expansion" are selected and comments are entered.
In this case, as shown in FIG. 14, “symbol”, “keyword”, “request”, and “Idea development” are added as new attributes to the contents of the data table in accordance with the idea generation work.
[0020]
Through the above process, data processing results can be obtained, and various analyzes can be added by editing work and idea generation work by analysts. To make this result into a report, it is usually necessary to recreate it again. It becomes necessary and becomes troublesome twice. Therefore, the present system includes a reporting unit 7 for reporting data and analysis results obtained in a series of operations.
[0021]
As shown in FIG. 15, the reporting unit 7 has a reporting format selection unit and a layout setting table for setting and recording the layout / character setting information. Therefore, when the analyst selects a reporting format, information necessary for the report is extracted from the contents of the data table 2, and various layout reports corresponding to the reporting format as shown in FIGS. 16 to 19 are obtained. Can do. 16 to 18 are examples in which the data and analysis results shown above are directly used in a report format, and FIG. 19 is an example in which a comment is added and a report is generated together with a graph display that summarizes the classification results.
[0022]
Embodiment 2. FIG.
FIG. 20 is an overall configuration diagram of the free opinion analysis system showing Embodiment 2 of the present invention, FIG. 21 is a diagram showing a specific implementation scene of cooperative work by this free opinion analysis system, and FIG. 22 is this free opinion analysis system. It is a figure which shows the update of the data table by this cooperative work.
[0023]
In the figure, the same reference numerals are given to the same or corresponding parts as those in the first embodiment, and the description will be omitted. Reference numeral 11 denotes a collaborative work input interface (the interface is hereinafter referred to as I / F), which indicates a plurality of input I / Fs such as a pointing device, a mouse, and a keyboard. Reference numeral 81 denotes a cooperative work output I / F, which indicates a plurality of output I / Fs such as a display and a projector.
[0024]
Next, the operation will be described.
FIG. 21 shows an example of a conference as an example of cooperative work. Here, one large-screen projector is provided as the output I / F 11 for cooperative work, and the editing screen, editing results, reports, etc. are displayed on the large-screen projector, and all analysts' viewpoints are directed to the large-screen projector. By matching, the knowledge of the analysis content is shared, and as a collaborative work input I / F 81, the conference moderator examines and edits each display content using the pointing device and other analysts using the mouse and keyboard. By doing so, efficient collaborative work is realized.
[0025]
In addition, as shown in FIG. 22, the information edited and input by the collaborative work I / F is registered in advance for each device, thereby identifying the input person and automatically analyzing it. Stored in the data table as an attribute of "user". Needless to say, information can be added to the report output by the reporting unit 7 as the attribute of the “analyzer”.
[0026]
Embodiment 3 FIG.
FIG. 23 is an overall configuration diagram of the free opinion analysis system showing Embodiment 3 of the present invention, and FIG. 24 is a diagram showing a cooperative work output I / F of this free opinion analysis system. In the figure, the same or corresponding parts as those in the first and second embodiments are denoted by the same reference numerals, and description thereof is omitted. Reference numeral 9 denotes an I / F for remote location cooperative work using a network or the like, and shows, for example, an environment for connection to the Internet.
[0027]
Next, the operation will be described.
Even if the analysts are at different locations, the remote window cooperative work I / F 9 is used to display a multi-window screen on each analyst's display as shown in FIG. This main window is a screen common to all analysts, and each work window is a screen that can be controlled by each analyst. Therefore, each analyst uses a mouse and a keyboard to examine and edit each display content, thereby realizing efficient collaborative work.
[0028]
In the free opinion analysis systems according to the first to third embodiments, the editing unit 5 and the idea idea support unit 6 are shown. However, only one of the editing unit 5 and the idea idea support unit 6 is provided. Also good.
[0029]
【The invention's effect】
Since the present invention is configured as described above, the following effects can be obtained.
[0032]
A data input unit for inputting at least data from a free description sentence, a data table unit for storing data input from the data input unit, a data processing unit for classifying data input from the data input unit, and An editing unit having a function of re-editing by humans on a result obtained by the data processing unit, and at least a comment input of a mark, a comment, or a character editing is performed on the processing result by the data processing unit or the editing unit. An idea idea support unit having a function; a reporting unit that processes the contents of each part in a report format; and an output unit that displays the contents of each part. The idea idea support unit includes a comment in which a plurality of comment titles are set. A title database, and for the processing result, the comment title database Comment title selected by an operation input from, and add as an attribute for classifying the data, a comment input, since the input as the attribute data of the attribute, the processing results are automatically analyzed by the data processing unit The editing department can make corrections and edits from the viewpoint of the analyst, obtain more accurate classification results, and discover new knowledge through KJ-style work that repeats opinion classification and division. It is also possible to record the free opinions and inspirational ideas and viewpoints that the analyst was interested in based on the processing results, thereby facilitating the idea generation.
[0033]
In addition, since it has an output interface for collaborative work in which the display by the output unit is shared by a plurality of people, while the analysts share the problems, the plurality of analysts are aware from different viewpoints, In the case of developing a product as well as obtaining enormous knowledge that cannot be obtained by a single work due to the synergistic effect of thinking out ideas and ideas, and further developing ideas based on the ideas that were developed The development direction (vector) can be matched between the development parties.
[0034]
In addition, since it has a collaborative input interface for inputting data to the data input unit by a plurality of people, a plurality of analysts come up with ideas, opinions, and ideas from different viewpoints, In addition, based on the idea that has been conceived, it is possible to obtain enormous knowledge that cannot be obtained by a single operation due to the synergistic effect of developing the idea further.
[0035]
In addition, since the remote input cooperative work interface that simultaneously cooperates with a plurality of people between remote locations in the data input unit and the output unit, even if not gathered in one place as in a normal meeting, It is possible to obtain the same effect as when gathered in one place.
[0036]
In addition, since a reporting unit for outputting the contents of each unit in a report format is provided, it is possible to automatically report analysis results including comments obtained in the analysis process.
[0037]
In addition, since it has a data mining section that aggregates and statistically analyzes the data processed by the data processing section, it performs various calculations such as arithmetic calculation, statistical analysis, multivariate analysis, and tables, graphs, and diagrams. It is possible to analyze from various angles.
[0038]
Further, the data handled by the data table section is any of free opinion data, option format category data, numerical data such as age, ranking data indicating rank, multiple selection multi-answer data, graphical data such as pictures and photographs, etc. Therefore, it is possible to have various data tables, to perform cross analysis using these various data, and to store attribute data such as hierarchy numbers, comments, etc. added during the analysis process, Ideas can be generated in a form that includes data, and reporting can also be performed.
[Brief description of the drawings]
1 is an overall configuration diagram of a free opinion analysis system showing Embodiment 1 of the present invention;
FIG. 2 is a diagram showing a specific example of data stored in a data table of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 3 is a diagram schematically showing the contents of a data table of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 4 is a diagram showing a data classification result of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 5 is a diagram showing a configuration of a data processing unit and a data table of processing results of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 6 is a configuration diagram of a data mining unit of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 7 is a diagram showing a processing result of a data mining unit of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 8 is a diagram showing another processing result of the data mining unit of the free opinion analysis system showing Embodiment 1 of the present invention;
FIG. 9 is a configuration diagram of an editing unit of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 10 is a diagram showing updating of a data table by editing work of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 11 is a diagram showing an edit screen of the free opinion analysis system showing Embodiment 1 of the present invention.
FIG. 12 is a configuration diagram of an idea idea support unit of the free opinion analysis system according to Embodiment 1 of the present invention.
FIG. 13 is a diagram showing a result of editing by an idea idea support unit of the free opinion analysis system according to Embodiment 1 of the present invention.
FIG. 14 is a diagram showing updating of a data table by an idea idea support unit of the free opinion analysis system according to Embodiment 1 of the present invention.
FIG. 15 is a configuration diagram of a reporting unit of the free opinion analysis system according to Embodiment 1 of the present invention.
FIG. 16 is a diagram showing an output example from the reporting unit of the free opinion analysis system showing Embodiment 1 of the present invention;
FIG. 17 is a diagram showing an output example from the reporting unit of the free opinion analysis system showing Embodiment 1 of the present invention;
FIG. 18 is a diagram showing an output example from the reporting unit of the free opinion analysis system showing Embodiment 1 of the present invention;
FIG. 19 is a diagram showing an output example from the reporting unit of the free opinion analysis system according to Embodiment 1 of the present invention;
FIG. 20 is an overall configuration diagram of a free opinion analysis system showing Embodiment 2 of the present invention.
FIG. 21 is a diagram showing a specific execution scene of cooperative work of the free opinion analysis system showing Embodiment 2 of the present invention;
FIG. 22 is a diagram showing the update of the data table by the cooperative work of the free opinion analysis system showing Embodiment 2 of the present invention.
FIG. 23 is an overall configuration diagram of a free opinion analysis system showing Embodiment 3 of the present invention.
FIG. 24 is a diagram illustrating an output I / F for cooperative work of the free opinion analysis system according to Embodiment 3 of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Data input part, 2 Data table, 3 Data processing part, 4 Data mining part, 5 Editing part, 6 Idea idea support part, 7 Reporting part, 8 Output part, 9 Remote area cooperation I / F, 11 Cooperation work Input I / F, 81 Collaborative output I / F.

Claims (6)

A data input unit for inputting at least data from a free description sentence, a data table unit for storing data input from the data input unit, a data processing unit for classifying data input from the data input unit, and An editing unit having a function of re-editing a result obtained by the data processing unit by a human, and at least a comment input of a mark, a comment, or a character editing is performed on the processing result by the data processing unit or the editing unit An idea idea support unit having a function, a reporting unit that processes the content of each unit into a report format, an output unit that displays the content of each unit,
Equipped with a,
The idea idea support unit has a comment title database in which a plurality of comment titles are set,
A comment title selected by an operation input from the comment title database is added to the processing result as an attribute for classifying the data, and the input comment is input as attribute data of the attribute. Free opinion analysis system. Free opinion analysis system of claim 1 Symbol mounting, characterized in that and a cooperative working output interface shared by several people the display by the output unit. Free opinion analysis system according to claim 1 or 2, characterized in that and a cooperative working input interface for data input in a plurality of human to the data input unit. Free opinion analysis system according to claim 1 to 3, wherein further comprising a, a remote collaborative work interface for collaborative work simultaneously on multiple human between remote locations to the data input unit and the output unit. Free opinion analysis system according to claim 1-4, wherein further comprising a data mining unit for various compiled and statistical analysis of the data that is processed by the data processing unit. The data handled by the data table section includes any of free opinion data, option format category data, numerical data such as age, ranking data indicating ranking, multiple selection multi-answer data, and graphical data such as pictures and photographs. The free opinion analysis system according to any one of claims 1 to 5 .

Images