JP6494687B2 - Evaluation apparatus, evaluation method, and evaluation program - Google Patents

Description

  The present invention relates to an evaluation apparatus, an evaluation method, and an evaluation program.

  Traditionally, information on brands used by companies and organizations (hereinafter referred to as “businesses”) is collected on various media such as TV, newspapers, magazines, and Web articles, and the level of brand exposure and brands in the world. Evaluation of the image is performed. By evaluating such brands, companies can use them to determine brand strategies.

  For example, Patent Document 1 collects information exposed to the media for a given brand, calculates the exposure of the brand to the media, the brand interest that represents public interest in the brand, and integrates them. Thus, a brand evaluation method for calculating a predetermined evaluation index has been proposed. In addition, brand evaluation methods using various indices are known (see Patent Documents 2 to 5).

JP-A-2015-95249 JP 2014-32636 A International Publication No. 2004-066175 International Publication No. 2004-055707 JP 2003-108738 A

  In the brand evaluation method described in Patent Document 1, information exposed to the media regarding a predetermined brand is acquired and an evaluation index is calculated. In this evaluation method, for example, a company that is the owner of the brand Couldn't evaluate the gap between the brand image they expect and the public reaction. Moreover, it was not possible to properly evaluate whether the information exposed to the media was positive or negative.

  In addition, in the conventional brand evaluation method, media such as television, newspapers, and magazines have been emphasized. However, in recent years, various media including social networking services (SNS) and the like that are prominent in the public reaction are comprehensively integrated. And it needs to be evaluated equally.

  The present invention has been made in consideration of such circumstances, and evaluates the gap between the brand image expected by the company and the reaction of the public (customer), and determines what the world is based on the activities performed by the company. It is an object to provide an evaluation apparatus, an evaluation method, and an evaluation program that can appropriately evaluate whether or not the reaction has occurred. In addition, when assessing the gap between brand image and public reaction, the present invention can appropriately evaluate the accuracy of information exposed to media, and can also assess the degree of risk of a company. An object is to provide an evaluation device, an evaluation method, and an evaluation program.

  (1): An extraction unit (for example, screening unit 12 in the embodiment) that extracts data related to a brand preset by a company or organization from data collected from an information medium, and the extraction unit extracts the data. In the generated data, a first index value indicating the appearance frequency of the first term indicating the content expected by the company or the organization and a second index value indicating the appearance frequency of the second term indicating the reaction of the consumer are calculated. It is an evaluation apparatus provided with the calculation part (for example, scoring part 18 in embodiment) to perform.

  (2): In (1), a first storage unit (for example, the mirror dictionary 36 in the embodiment) that stores the first term and a second storage unit (for example, the thermostat in the embodiment) that stores the second term. The calculation unit calculates the first index value based on the first term stored in the first storage unit, and stores the first index value stored in the second storage unit. The second index value is calculated based on two terms.

  (3): In (1) or (2), the calculation unit indicates positive contents for each of the data collected from the information medium for each of the first index value and the second index value. Or whether it indicates a negative content.

  (4): In any one of (1) to (3), the calculation unit may calculate the data including the first term and the data including the second term based on the type of the information medium. Weighting to calculate the first index value and the second index value.

  (5): In any one of (1) to (4), a generation unit that generates information that enables display of an image that compares the first index value and the second index value (for example, implementation) It further comprises a scoring part 18) in the form.

  (6): Extracting data related to a brand preset by a company or organization from data collected from an information medium, and indicating the content expected by the company or organization in the extracted data This is an evaluation method for calculating a first index value indicating the appearance frequency of one term and a second index value indicating the appearance frequency of a second term indicating a consumer reaction.

  (7): Causes a computer to extract data related to a brand preset by a company or organization from data collected from an information medium, and the content expected by the company or organization in the extracted data This is an evaluation program for calculating a first index value indicating the appearance frequency of the first term indicating, and a second index value indicating the appearance frequency of the second term indicating the reaction of the customer.

  (8): An extraction unit (for example, screening unit 12 in the embodiment) that extracts data related to a brand set in advance by a company or organization from data collected from an information medium, and the extraction unit extracts The first index value indicating the appearance frequency of the first term indicating the content expected by the company or the organization in the generated data, the second index value indicating the appearance frequency of the second term indicating the reaction of the consumer, An evaluation apparatus including a calculation unit (for example, a scoring unit 18 in the embodiment) that calculates a frequency of appearance of a third term indicating a risk of a company or a group and a third index value based on the type of the information medium.

  (9): In (8), the first storage unit storing the first term (for example, the mirror dictionary 36 in the embodiment) and the second storage unit storing the second term (for example, the thermo in the embodiment). A dictionary 38) and a third storage unit (for example, the risk dictionary 40 in the embodiment) storing the third term, and the calculation unit uses the first term stored in the first storage unit. Based on the second term stored in the second storage unit, the second index value is calculated based on the second term stored in the second storage unit, and the third index stored in the third storage unit Based on the terminology, the third index value is calculated.

  (10): In (8) or (9), the calculation unit, for each of the data collected from the information medium, for each of the first index value, the second index value, and the third index value, It is determined whether it indicates a positive content or a negative content.

  (11): In any one of (8) to (10), the calculation unit may calculate the data including the first term and the data including the second term based on the type of the information medium. Weighting to calculate the first index value and the second index value.

  (12): In any one of (8) to (11), the calculation unit calculates an appearance frequency of the third term for each information medium, and the calculated third term for each information medium The third index value is calculated on the basis of the frequency of occurrence and the threshold value set for each information medium.

  (13): In any one of (8) to (12), information that enables display of an image that compares the first index value, the second index value, and the third index value is generated. A generating unit (for example, the scoring unit 18 in the embodiment).

  (14): Extracting data related to a brand preset by a company or organization from data collected from an information medium, and indicating the contents expected by the company or organization in the extracted data A first index value indicating an appearance frequency of one term, a second index value indicating an appearance frequency of a second term indicating a reaction of a customer, an appearance frequency of a third term indicating a risk of the company or the organization, and the information This is an evaluation method for calculating a third index value based on the type of medium.

  (15): Causes a computer to extract data related to a brand set in advance by a company or organization from data collected from an information medium, and contents expected by the company or organization in the extracted data The first index value indicating the appearance frequency of the first term indicating, the second index value indicating the appearance frequency of the second term indicating the reaction of the customer, and the appearance frequency of the third term indicating the risk of the company or group And an evaluation program for calculating a third index value based on the type of the information medium.

  According to (1), (2), (5), (6), (7), (8), (9), (13), (14), (15), a brand expected by a company or organization It is possible to evaluate the gap between the image of the public and the reaction of the public (customer), and to appropriately evaluate how the public responded to the activities carried out by companies or organizations. In addition, when evaluating the gap between brand image and public reaction, it is possible to evaluate appropriately including the certainty of information exposed to the media, and also to assess the degree of risk of the company.

  According to (3) and (10), it is possible to grasp whether each piece of data collected from the information medium indicates a positive content or a negative content.

  According to (4), (11), and (12), data collected from various information media can be evaluated comprehensively and equally.

It is a functional block diagram which shows an example of the evaluation apparatus in embodiment. It is a figure which shows an example of the dictionary data memorize | stored in the mirror dictionary in embodiment. It is a figure which shows an example of the dictionary data memorize | stored in the thermo dictionary in embodiment. It is a figure which shows an example of the dictionary data memorize | stored in the risk dictionary in embodiment. It is a figure which shows the other example of the dictionary data memorize | stored in the risk dictionary in embodiment. It is a flowchart which shows an example of the flow of a process of the evaluation apparatus in embodiment. It is a flowchart which shows an example of the flow of the tagging process of the evaluation apparatus in embodiment. It is a figure which shows an example of the process target data in which the tagging process was performed in embodiment. It is a flowchart which shows an example of the flow of the scoring process of the evaluation apparatus in embodiment. It is a figure which shows an example of the scoring process of the evaluation apparatus in embodiment. It is a figure which shows an example of the time series data of the mirror score and thermo score with respect to the brand A in embodiment. It is a figure which shows an example of the time-sequential data of the mirror score and the thermo score with respect to the brand B in embodiment. It is a figure which shows an example of the mirror score classified by classification with respect to the brand A in embodiment. It is a figure which shows an example of the thermo score classified by classification with respect to the brand A in embodiment. It is a figure which shows an example of the time-sequential data of the thermoscore regarding the brand A in embodiment, a mirror score, a risk value, and the number of cases of processing target data. It is a figure which shows another example of the risk level determination process of the evaluation apparatus in embodiment.

  Hereinafter, an evaluation apparatus, an evaluation method, and an evaluation program according to some embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a functional block diagram illustrating an example of an evaluation apparatus 1 according to an embodiment of the present invention. The evaluation device 1 collects and analyzes data to be processed (hereinafter referred to as “processing target data”) from the medium M (information medium), thereby performing evaluation on the brand to be evaluated. The media M includes, for example, an SNS such as a television, a newspaper, a magazine, a web article, a web log, and a short text posting service, and a shareholder report.

  The evaluation device 1 calculates three values, “mirror score (first index value)”, “thermo score (second index value)”, and “risk value (third index value)” as evaluation indexes related to the brand. . “Mirror score” refers to how the company thinks about the brand in relation to the definition of how the company that owns the brand to be evaluated wants to see the brand. It is an index to represent. The “mirror score” indicates the appearance frequency of a term (first term) indicating contents expected by a company. By calculating the mirror score, it is possible to grasp the gap between the brand image expected by the company and the public reaction.

  “Thermo-score” is an index that indicates whether the public's emotions could be enhanced by the activities of the company that owns the brand to be evaluated. “Thermo-score” indicates the appearance frequency of a term (second term) indicating the reaction of the world (customer). By calculating this thermoscore, it is possible to grasp whether the public's feelings about the brand to be evaluated could be increased or the support of the public could be obtained. The “risk value” is an index indicating the degree of risk that has occurred with respect to the brand to be evaluated. The “risk value” is calculated based on the appearance frequency of a term (third term) indicating the risk of a company or an organization and the type of media M (information medium).

  The evaluation apparatus 1 includes, for example, a collection unit 10, a screening unit 12 (extraction unit), an analysis unit 14, a tagging unit 16, a scoring unit 18 (calculation unit, generation unit), a display unit 20, A storage unit 22 and a dictionary DB 24 are provided. The dictionary DB 24 includes, for example, a corporate dictionary 30, a topic dictionary 32, a sensitivity dictionary 34, a mirror dictionary 36 (first storage unit), a thermo dictionary 38 (second storage unit), and a risk dictionary 40 (third storage). Part).

  The collection unit 10 collects processing target data from the medium M and stores it in the storage unit 22. The collection unit 10 collects processing target data via the Internet N, for example. The collection unit 10 collects data to be processed at a predetermined timing such as daily or weekly. The media M that is the collection source may be determined in advance by the user of the evaluation device 1. In addition, the collection unit 10 may perform a crawling process for periodically collecting documents on the Internet including a predefined character string. Note that the evaluation device 1 may include a reception unit (not shown) that receives an input from the operator P, and the collection unit 10 may collect the processing target data input to the reception unit.

  The screening unit 12 excludes the processing target data not related to the brand to be evaluated from the processing target data collected by the collection unit 10. The screening unit 12 extracts data related to a brand preset by a company or organization from data collected from the medium M (information medium). For example, the screening unit 12 reads dictionary data stored in the corporate dictionary 30. This dictionary data becomes a brand to be evaluated. And the screening part 12 excludes the process target data containing the homonym of this brand. In addition, for example, the screening unit 12 excludes the processing target data acquired from the specific URL defined in advance, excludes the processing target data acquired from the web log and the SNS of the predetermined specific ID, and the same text If there are a large number of data including, it may be excluded as abnormal data.

  The analysis unit 14 performs a morpheme analysis process that divides the text included in the processing target data into word levels and associates brands having the same meaning. For example, the analysis unit 14 associates alphabet notation indicating the same brand, kanji notation, katakana notation, hiragana notation, and the like, and processes the data as data indicating the same brand. Moreover, the analysis part 14 may process as data which shows the same brand, even if there exists a difference in a capital letter, a small letter, and the mixed character of a capital letter and a small letter regarding alphabet description. Further, the analysis unit 14 may process text including an error (such as a Kanji conversion error) as data indicating a brand to be evaluated. Thereby, it is possible to correct the fluctuation of the notation.

  In addition, the analysis unit 14 grasps the relationship between the dependency source and the dependency destination of the word included in the text, and the strength of expression of the word included in the text, multiple negation, affirmative question, dependency, comparison, dialect Performs parsing processing for interpretation. Interpreting the strength of the expression means, for example, that “very” in the expression “A product is very good” and “slight” in the expression “A product is slightly better” It is to be interpreted as a strong expression. Also, to interpret multiple negation is to correctly interpret, for example, an expression including a double negation that “A product is not bad” as a positive expression.

  In addition, to interpret a positive question is, for example, to interpret a question sentence expressed with a positive intention “A product is a good product?” As a positive expression. Interpretation of dependency is, for example, by correctly interpreting its meaning even when the position of the modifier is reversed, such as the expression “Is the product A good?” is there. This “Is product good?” Is interpreted as a positive expression.

  Also, to interpret the comparison, it is intended that the A product is worse for the expression that compares the current model of the A product with the previous model that “A product was better than the previous model”. It is to judge and interpret as a negative expression. Interpreting a dialect means, for example, that the dialect “Menkoi” in the expression “A product is a noodle product” is interpreted as the standard word “cute” and its meaning is interpreted correctly. . This “A product is a noodle product” is interpreted as a positive expression. By performing the parsing process as described above, the accuracy of semantic understanding can be increased.

  The tagging unit 16 refers to the company dictionary 30, the topic dictionary 32, the sensitivity dictionary 34, the mirror dictionary 36, the thermo dictionary 38, and the risk dictionary 40, and applies the text included in each processing target data. Then, “company tag”, “topic tag”, “sensibility tag”, “mirror tag”, “thermo tag”, and “risk tag” are tagged. Details of the tagging process will be described later.

  The scoring unit 18 performs a scoring process for calculating a mirror score, a thermo score, and a risk value based on the processing target data subjected to the tagging process by the tagging unit 16. For example, the scoring unit 18 calculates a mirror score for a character string tagged with a combination of “company tag”, “topic tag”, and “mirror tag” in the processing target data. For example, the scoring unit 18 calculates a thermo score for a character string tagged with a combination of “company tag”, “topic tag”, and “thermo tag” in the processing target data. For example, the scoring unit 18 calculates a risk value for a character string tagged with a combination of “company tag”, “topic tag”, and “risk tag” in the processing target data. For example, the scoring unit 18 calculates a risk value for a character string tagged with a combination of “company tag” and “risk tag” in the processing target data.

  Part or all of the collection unit 10, the screening unit 12, the analysis unit 14, the tagging unit 16, and the scoring unit 18 is realized by a processor (computer) executing a program (software). Some or all of these may be realized by hardware such as LSI (Large Scale Integration) or ASIC (Application Specific Integrated Circuit), or may be realized by a combination of software and hardware.

  The display unit 20 displays the mirror score, thermo score, and risk value calculated by the scoring unit 18. The user of the evaluation device 1 can evaluate the target brand by confirming the mirror score, the thermo score, and the risk value displayed on the display unit 20. The display unit 20 is, for example, a liquid crystal display or an organic EL (Electroluminescence) display device.

  The storage unit 22 stores the processing target data collected by the collection unit 10. The storage unit 22 stores, for example, the type of collection source media and processing target data in association with each other.

  The dictionary DB 24 stores dictionary data used in various processes performed by the screening unit 12, the analysis unit 14, the tagging unit 16, and the scoring unit 18. The storage unit 22 and the dictionary DB 24 are realized by a ROM (Read Only Memory), a RAM (Random Access Memory), a HDD (Hard Disk Drive), a flash memory, or the like.

  The enterprise dictionary 30 is a brand owned by the enterprise and stores dictionary data of a brand to be evaluated. The company dictionary 30 stores terms that characterize the company, such as a company name, a product name, a service name, and the name of the company president.

  The topic dictionary 32 stores dictionary data of topics that are assumed to be topics in the world. For example, when information such as a news release is announced to the public regarding a brand to be evaluated, the topic dictionary 32 stores a topic described in this news release. The topic dictionary 32 stores, for example, terms associated with the brand to be evaluated, such as “new model”, “release start”, “announcement”, “world announcement”.

  The sensitivity dictionary 34 stores dictionary data used for semantic analysis of the processing target data. The sensitivity dictionary 34 stores dictionary data necessary for positive and negative determination processing of processing target data performed by the scoring unit 18.

  The mirror dictionary 36 stores dictionary data that defines how a brand of a company wants to be seen in the world and what words the brand is desired to represent. FIG. 2 is a diagram showing an example of dictionary data stored in the mirror dictionary 36 in the present embodiment. The mirror dictionary 36 stores a term indicating at least one major category, a term indicating at least one middle category associated with each major category, and a keyword associated with each major category. This keyword is used as dictionary data.

  For example, in the example shown in FIG. 2, “trust” is stored as major classification 1, “safety” is stored as middle classification 1-1 associated with major classification 1, and “high quality” is classified as middle classification 1-2. Is stored. Further, “safety”, “reliable”, and “trust” are stored as keywords associated with the middle class 1-1, and “high quality”, “safety”, and “not broken” are keywords associated with the middle class 1-2. Is stored. The mirror dictionary 36 does not need to have a data structure of major classification, middle classification, and keyword, and may define only a major classification, a more detailed classification, or a classification definition. Instead, only the keyword may be stored.

  The thermo dictionary 38 stores dictionary data for evaluating whether the public's emotions can be enhanced by the activities of the company that owns the brand to be evaluated. FIG. 3 is a diagram showing an example of dictionary data stored in the thermo dictionary 38 in the present embodiment. In the example illustrated in FIG. 3, the thermo dictionary 38 stores “relief”, “satisfaction”, “like”, and “expectation” as terms representing positive feelings of the world. In addition, the thermo dictionary 38 stores terms opposite to the above-mentioned positive emotional terms such as “anxiety”, “dissatisfaction”, “dislike”, and “disappointment” as terms representing negative emotions. ing. The thermo dictionary 38 may store a keyword associated with a term representing the emotion. The thermo-dictionary 38 may store, for example, “check now” and “looking forward” as keywords associated with “expectation”.

  The risk dictionary 40 stores dictionary data indicating a risk that a company should grasp for a brand to be evaluated. FIG. 4A is a diagram showing an example of dictionary data stored in the risk dictionary 40 in the present embodiment. FIG. 4B is a diagram showing another example of dictionary data stored in the risk dictionary 40 in the present embodiment. The risk dictionary 40 includes at least one major category term, at least one middle category term associated with each major category, at least one minor category term associated with each major category, The keyword associated with the minor classification is stored. This keyword is used as dictionary data.

  For example, in the example shown in FIG. 4A and FIG. 4B, “company origin” is stored as major category 1, “management quality” is stored as middle category 1-1 associated with major category 1, and middle category 1— 2 stores “product quality”. Furthermore, “performance” is stored as the minor category 1-1-1 associated with the middle category 1-1, and “personnel” is stored as the minor category 1-1-2. Further, “deterioration in business performance”, “price drop in stock price”, and “injustice” are stored as keywords associated with the minor category 1-1-1, and “restructuring”, “ "Hard work" and "Dismissal" are remembered. The risk dictionary 40 does not need to have a data structure of major classification, middle classification, minor classification, and keyword, and may set only the major classification or only the major classification and the middle classification, or define a finer classification. Alternatively, only the keywords may be stored without defining the classification.

  Next, operation | movement of the evaluation apparatus 1 in this embodiment is demonstrated. FIG. 5 is a flowchart showing an example of the processing flow of the evaluation apparatus 1 in the present embodiment.

  First, the collection unit 10 collects processing target data from the medium M and stores it in the storage unit 22 (step S101). The collection unit 10 collects processing target data via the Internet N, for example. The collection unit 10 collects data to be processed at a predetermined timing such as daily or weekly. Further, the evaluation device 1 may include a reception unit (not shown) that receives input from the operator P, and the collection unit 10 may collect the processing target data input to the reception unit.

  Next, the screening unit 12 reads the processing target data stored in the storage unit 22 and performs a screening process for excluding data not related to the brand to be evaluated (step S103). For example, the screening unit 12 reads dictionary data (brand to be evaluated) from the corporate dictionary 30 and excludes data including homonyms of this brand.

  Next, the analysis unit 14 divides the text included in the processing target data into word levels, and performs a morphological analysis process that associates brands having the same meaning (step S105). For example, the analysis unit 14 associates alphabet notation indicating the same brand, kanji notation, katakana notation, hiragana notation, and the like, and processes the data as the same brand to correct the shake of the notation.

  Next, the analysis unit 14 grasps the relationship between the dependency source and the dependency destination of the word included in the text, and expresses the strength of the expression of the word included in the text, multiple negation, affirmative question, dependency, comparison, dialect A parsing process for interpreting is performed (step S107).

  Next, the tagging unit 16 applies “company tag”, “topic tag”, “sensitivity tag”, “mirror tag”, “thermo tag”, and “risk tag” to the text included in each processing target data. Tagging is performed (step S109).

  FIG. 6 is a flowchart showing an example of the flow of the tagging process (step S109) of the evaluation apparatus 1 in the present embodiment. First, the tagging unit 16 extracts one data from the processing target data subjected to the syntax analysis processing by the analysis unit 14 (step S201).

  Next, the tagging unit 16 applies “company tag”, “topic tag”, “sensitivity tag”, “mirror tag”, “thermo tag”, and “risk tag” to the text included in each processing target data. Tagging is performed (step S203). When the character string of the dictionary data stored in the company dictionary 30 is included in the processing target data, the tagging unit 16 assigns a “company tag” to the character string in the processing target data. When the character string of the dictionary data stored in the topic dictionary 32 is included in the processing target data, the tagging unit 16 assigns a “topic tag” to the character string in the processing target data. When the character string of the dictionary data stored in the sensitivity dictionary 34 is included in the processing target data, the tagging unit 16 assigns a “sensitivity tag” to the character string in the processing target data. If the character string of the dictionary data stored in the mirror dictionary 36 is included in the processing target data, the tagging unit 16 assigns a “mirror tag” to the character string in the processing target data. When the character string of the dictionary data stored in the thermo dictionary 38 is included in the processing target data, the tagging unit 16 adds a “thermo tag” to the character string in the processing target data. When the character string of the dictionary data stored in the risk dictionary 40 is included in the processing target data, the tagging unit 16 assigns a “risk tag” to the character string in the processing target data.

  FIG. 7 is a diagram illustrating an example of processing target data on which tagging processing has been performed in the present embodiment. In the example shown in FIG. 7, “company tag” is assigned to the character strings “Company A” and “product A”, “topic tag” is assigned to the character string “start sales”, and “original” "Mirror tag" is assigned to the character string "Self", "Thermo tag" and "Kansei tag" are assigned to the character string "Expectation", and "Risk tag" is assigned to the character string "Failure" Has been granted.

  Next, the tagging unit 16 determines whether or not the tagging process for all the processing target data has been completed (step S205). If the tagging unit 16 determines that tagging processing has not been completed for all processing target data, the tagging unit 16 extracts processing target data that has not been tagged and performs tagging processing. On the other hand, the tagging unit 16 completes the tagging process when determining that the tagging process for all the processing target data is completed.

  Next, the scoring unit 18 performs a scoring process for calculating a mirror score, a thermo score, and a risk value based on the processing target data tagged by the tagging unit 16 (step S111).

  FIG. 8 is a flowchart illustrating an example of the flow of the scoring process (step S111) of the evaluation device 1 in the present embodiment. First, the scoring unit 18 extracts one data from the processing target data subjected to the tagging process by the tagging unit 16 (step S301).

  Next, the scoring unit 18 targets the processing target data tagged with the combination of “corporate tag”, “topic tag”, and “mirror tag” for each processing target data (hereinafter referred to as “individual mirror”). (Score) is calculated (step S303). For example, the scoring unit 18 counts the number of “mirror tags” and calculates an individual mirror score with one “mirror tag” as one point.

  Next, the scoring unit 18 targets the processing target data tagged with a combination of “company tag”, “topic tag”, and “thermo tag” for each processing target data (hereinafter referred to as “individual thermostat”). (Score) is calculated (step S305). For example, the scoring unit 18 counts the number of “thermo tags” and calculates an individual thermo score with one “thermo tag” as one point.

  FIG. 9 is a diagram illustrating an example of the scoring process of the evaluation device 1 in the present embodiment. In the example shown in FIG. 9, “processing target data 1” includes a character string to which “topic tag” “new model” is assigned, a character string to which “company tag” “product A” is assigned, and “ It includes a character string to which “mirror tag (major classification“ Yorokobi ”)” “It was cool” and a character string to which “thermo tag (keyword associated with“ expectation ”)” “check” is added. In this case, the scoring unit 18 calculates the individual mirror score as “1.0” by the combination of “Product A”, “New” and “Yorokobi”. Further, the scoring unit 18 calculates the individual thermo score as “1.0” by the combination of “product A”, “new model”, and “expectation”.

  “Processing data 2” includes a character string with a “topic tag” of “new type”, a character string with a “mirror tag (major category“ Yorokobi ”)” of “cool”, And a character string provided with a “thermo tag (keyword associated with“ expectation ”)”. In this case, since there is no character string to which the “company tag” is assigned, the scoring unit 18 sets the individual mirror score and the individual thermo score to “0.0”.

  “Processing data 3” includes a character string to which “topic tag” “new model” is assigned, a character string to which “company tag” “product A” and “company A” is assigned, and “it was cool”. And a character string to which “thermo tag (keyword associated with“ expectation ”)” is assigned. That is, “processing target data 3” includes two character strings “product A” and “company A” to which “company tag” is assigned. In this case, the scoring unit 18 calculates an individual mirror score and an individual thermo score for each of the two character strings.

  Here, “processing object data 3” includes a positive expression “exhibition was very cool”, but this may not express product A or company A positively. In other words, the expression “exhibition was really cool” can be interpreted as a positive expression about the method of “exhibition”. In this case, the scoring unit 18 calculates an individual mirror score and an individual thermo score that are lower than the processing target data 1 that directly expresses the product A positively. For example, the scoring unit 18 calculates an individual mirror score of “0.5” by combining “Product A”, “New Model”, and “Yorokobi”, and “Company A”, “New Model”, and “Yorokobi” The individual mirror score is calculated as “0.5” with the combination of In addition, the scoring unit 18 calculates an individual thermo score of “0.5” by combining “Product A”, “New Model”, and “Expectation”, and “Company A”, “New Model”, and “Expectation” The individual thermo score is calculated as “0.5” with the combination of

  Next, the scoring unit 18 targets the processing target data tagged with a combination of “company tag”, “topic tag”, and “risk tag” (hereinafter, “individual risk data”). (Referred to as “risk value”) (step S307). For example, the scoring unit 18 counts the number of “risk tags” and performs scoring with one “risk tag” as one point.

  Next, the scoring unit 18 determines whether or not the individual scoring processing for all the processing target data has been completed (step S309). When the scoring unit 18 determines that the individual scoring processing for all the processing target data has not been completed, the scoring unit 18 extracts the processing target data that has not been subjected to the individual scoring processing and performs the individual scoring processing. .

  On the other hand, when the scoring unit 18 determines that the individual scoring processing for all the processing target data is completed, the scoring unit 18 performs weighting processing based on the type of collection source of each processing target data (step S311). The scoring unit 18 weights the data including the terms stored in the mirror dictionary 36 and the data including the terms stored in the thermo dictionary 38 based on the type of the medium M (information medium). Calculate the mirror score and thermo score. For example, based on past statistical data, the average utterance amount of a brand to be evaluated per day for each medium is calculated, and weighting is performed so that the influence of all the media is uniform. For example, when the weight of the newspaper is set to the reference value “1”, a weight “2” larger than that of the newspaper is set for a television with a small amount of speech compared to the newspaper. In addition, for an SNS that has a larger amount of speech than a newspaper, a smaller weight “0.1” than that of a newspaper is set.

  Next, the scoring unit 18 determines whether the content of the processing target data is a positive expression or a negative expression based on the “Kansei tag” given to the processing target data. Is performed (step S313). The scoring unit 18 determines whether each of the processing target data indicates a positive (positive) content or a negative (negative) content for each mirror score and thermo score. To do.

  For example, the scoring unit 18 counts the number of places where the number of positive expressions (the number of sensitivity tags of positive expressions) shows a negative expression (sensitivity of negative expressions) in one processing target data. If the number is larger than the number of tags), it is determined that the processing target data as a whole is a positive expression. Also, the scoring unit 18 determines that the processing target data is negative as a whole when the number of places showing a negative expression is larger than the number of places showing a positive expression in one processing target data. It is determined that it is an expression. The scoring unit 18 determines that the processing target data as a whole is the same when the number of locations showing positive expressions is the same as the number of locations showing negative expressions in one processing target data. Judged as neutral expression. In addition, the scoring unit 18 determines that the processing target data to which the “sensitivity tag” is not attached is emotionless.

  Next, the scoring unit 18 calculates a final thermo score and mirror score based on the processing target data subjected to the emotion determination process (step S315). For example, the scoring unit 18 adds, for each brand to be evaluated, the sum of the individual mirror scores of the processing target data determined to be a positive expression and the individual mirror score of the processing target data determined to be a negative expression. Calculate the sum of. Further, the scoring unit 18 subtracts the sum of the individual mirror scores of the processing target data determined to be negative expression from the total of the individual mirror scores of the processing target data determined to be positive expression, Calculate as the final mirror score.

  Further, for example, the scoring unit 18 adds, for each brand to be evaluated, the sum of the individual thermo scores of the processing target data determined to be a positive expression and the individual processing target data determined to be a negative expression. Calculate the total thermoscore. Further, the scoring unit 18 subtracts the sum of the individual thermo scores of the processing target data determined to be negative expression from the total of the individual thermo scores of the processing target data determined to be positive expression, Calculate as the final thermoscore. The scoring unit 18 also stores the calculated mirror score and thermo score in the storage unit 22.

  Next, the scoring unit 18 calculates a final risk value based on the processing target data subjected to the emotion determination process, and determines the risk level (step S317). In the determination of the risk level, for example, “high”, “medium”, and “low” of the risk are determined based on a predetermined threshold. In this case, a threshold may be set for each classification in the risk dictionary shown in FIGS. 4A and 4B. For example, even if the risk values are the same, if the “failure” in the minor category 1-2-1 is considered to be more serious than the “performance” in the minor category 1-1-1, The threshold value of “failure” in the minor category 1-2-1 is set lower than the “performance” in the minor category 1-1-1. Thus, the scoring process is completed.

  Next, the scoring unit 18 performs an evaluation result output process for causing the display unit 20 to display the calculated mirror score, thermo score, risk value, and the like (step S113). The scoring unit 18 generates information that enables display of an image that compares the mirror score, the thermo score, and the risk value. FIG. 10 is a diagram illustrating an example of time-series data of the mirror score and the thermo score for the brand A in the present embodiment. FIG. 11 is a diagram illustrating an example of time-series data of the mirror score and the thermo score for the brand B in the present embodiment. In the example shown in FIG. 10, on July 20, 2016, the values of the mirror score and the thermoscore are greatly reduced and a peak occurs, and on November 1, 2016, the values of the mirror score and the thermoscore are significantly increased. It can be seen that a peak occurs and the values of the mirror score and the thermo score are around 0 in other periods. Moreover, in the example shown in FIG. 11, although the peak of a mirror score and a thermoscore has generate | occur | produced frequently, it turns out that the value of each peak is smaller than the value of the peak in the brand A shown in FIG.

  FIG. 12 is a diagram illustrating an example of the classification-specific mirror score for the brand A in the present embodiment. In the example shown in FIG. 12, from May 1, 2016 to November 1, every major category 1 (trust), major category 2 (new), major category 3 (challenge) stored in the mirror dictionary shown in FIG. The average value per day of the mirror score of the processing target data determined to be a positive expression and the average value of the mirror score of the processing target data determined to be a negative expression per day. Shown as a bar graph. The mirror score of the processing target data determined to be a negative expression is shown as a negative value. Further, a line graph shows a value obtained by subtracting the average value of the mirror scores of the processing target data determined to be negative expression from the average value of the mirror scores of the processing target data determined to be positive expression. Thereby, the user of the evaluation apparatus 1 can confirm the mirror score classified by classification.

  FIG. 13 is a diagram illustrating an example of the classification-specific thermo score for the brand A in the present embodiment. In the example shown in FIG. 13, for each “Relief”, “Satisfaction”, “Like” and “Expectation” stored in the thermo dictionary shown in FIG. The bar graph shows the average daily thermoscore value of the processing target data determined to be positive and the average daily thermoscore value of the processing target data determined to be negative. . In addition, a line graph shows a value obtained by subtracting the average value of the thermo score of the processing target data determined to be negative expression from the average value of the thermo score of the processing target data determined to be positive expression. Thereby, the user of the evaluation apparatus 1 can confirm the thermo score according to classification.

  FIG. 14 is a diagram illustrating an example of time-series data of the thermoscore, the mirror score, the risk value, and the number of processing target data regarding the brand A in the present embodiment. FIG. 14 shows that the average value of the thermo score for each specific day (November 1, 2016) by category is “71” and the average value of the mirror score by category is “65”. Has been. In FIG. 14, regarding the risk level of the risk value, the number of cases determined as “high” is “2”, the number of cases determined as “medium” is “11”, and is determined as “low”. It is shown that the number of cases is “34”. Moreover, in FIG. 14, the number of the process target data regarding the brand A for the past 30 days is indicated by a line graph. Thereby, the user of the evaluation apparatus 1 can confirm the thermo score, the mirror score, and the risk level regarding the brand A. FIG. 14 shows an example in which the thermo score, the mirror score, and the risk value for a specific day are displayed. However, the thermo score, the mirror score, and the risk value for an arbitrary period such as a week unit or a month unit are shown. May be displayed.

  FIG. 15 is a diagram illustrating another example of the risk level determination process of the evaluation device 1 according to the present embodiment. First, the scoring unit 18 determines the validity of the tag combination assigned to each piece of processing target data (step S401). For example, the scoring unit 18 determines whether a tag is given to each piece of processing target data by a combination of “company tag” and “risk tag”. The scoring unit 18 uses the processing target data as a risk level determination target when a tag is given to the processing target data in a combination of “company tag” and “risk tag”. On the other hand, the scoring unit 18 excludes the processing target data from the risk level determination target when the processing target data is not provided with a combination of “company tag” and “risk tag”.

  The scoring unit 18 determines that the processing target data is a positive expression by the emotion determination process even if a tag is given to the processing target data by a combination of “company tag” and “risk tag”. If determined, the processing target data is excluded from the risk level determination target.

  Next, the scoring unit 18 aggregates risk values for each medium M for each of the character strings associated with the “company tag” with respect to the processing target data that is subject to risk level determination (step S403). ). Next, the scoring unit 18 determines, for example, a risk level of “large”, “medium”, and “small” for each of the character strings associated with the “company tag” with reference to a predetermined threshold ( Step S405). This threshold is set for each category in the risk dictionary shown in FIGS. 4A and 4B for each medium M.

  For example, regarding the “defects” of the subcategory 1-2-1 shown in FIG. 4A and FIG. To the first threshold corresponding to the risk level “small”, the second threshold corresponding to the risk level “medium” (first threshold <second threshold), and the third threshold corresponding to the risk level “large”. A threshold value (second threshold value <third threshold value) is set. Here, since “Web article”, “Web log”, and “SNS” have a larger amount of statements compared to “Newspaper” and “TV”, “Web article”, “Web log”, and “SNS”. May be set to a higher threshold than “newspaper” and “television”. For example, the scoring unit 18 determines that the risk level is “large” when the aggregated risk value is equal to or greater than the third threshold, and the risk level is equal to or greater than the second threshold and less than the third threshold. If it is determined to be “medium” and is not less than the first threshold and less than the second threshold, it is determined that the risk level is “small”. The scoring unit 18 may determine that there is no risk when the aggregated risk value is less than the first threshold.

  Next, the scoring unit 18 totals the risk levels determined for each medium M for each of the character strings associated with the “company tag” (step S407). For example, the scoring unit 18 uses a numerical value associated with each of the risk levels “large”, “medium”, and “small”. As numerical values associated with each of the risk levels “Large”, “Medium”, and “Small”, for example, values that decrease in the order of “Large”, “Medium”, and “Small” risk levels are set. The scoring unit 18 calculates, for each of the character strings associated with the “company tag”, the sum (total value) of the numerical values associated with the risk level determined for each medium.

  As described above, the scoring unit 18 totals the risk level determination results using cross media, and determines the final risk level based on the total value. For example, the scoring unit 18 sets the final risk levels “high”, “medium”, “low”, “safe” for each of the character strings associated with the “company tag” with reference to a predetermined threshold. Is determined.

  The scoring unit 18 may generate a heat map that expresses the final risk level determined by the scoring unit 18 in color. In this heat map, for example, character strings (character strings of dictionary data stored in the company dictionary 30) associated with “company tags” are arranged in the column direction, and classifications in the risk dictionary are arranged in the row direction. The final risk level determined by the scoring unit 18 is one square of this heat map.

  The calculated mirror score, thermo score, and risk value may be mapped in a three-dimensional space, and the evaluation value may be calculated by a support vector machine.

  According to the evaluation apparatus of the present embodiment described above, the gap between the brand image expected by the company and the reaction of the public is evaluated, and an appropriate evaluation is made as to how the public reacted to the activity performed by the company. Can do. In addition, it is possible to evaluate processing target data collected from various media in a comprehensive and equal manner. In addition, when evaluating the gap between brand image and public reaction, it is possible to evaluate appropriately including the certainty of information exposed to the media, and also to assess the degree of risk of the company.

  As mentioned above, although the form for implementing this invention was demonstrated using embodiment, this invention is not limited to such embodiment at all, and various deformation | transformation and substitution are within the range which does not deviate from the summary of this invention. Can be added.

DESCRIPTION OF SYMBOLS 1 ... Evaluation apparatus, 10 ... Collection part, 12 ... Screening part (extraction part), 14 ... Analysis part, 16 ... Tagging part, 18 ... Scoring part (calculation part, production | generation part), 20 ... Display part, 22 ... Storage unit, 24 ... Dictionary DB, 30 ... Corporate dictionary, 32 ... Topic dictionary, 34 ... Sensitivity dictionary, 36 ... Mirror dictionary (first storage unit), 38 ... Thermo dictionary (second storage unit), 40 ... Risk dictionary ( Third storage unit)

Claims (15)

An extractor for extracting data related to a brand preset by a company or organization from data collected from an information medium;
In the data extracted by the extraction unit, a first index value indicating the appearance frequency of the first term indicating the content expected by the company or the organization, and a second value indicating the appearance frequency of the second term indicating the reaction of the customer. An evaluation device comprising: a calculation unit that calculates an index value. A first storage unit storing the first term;
A second storage unit storing the second term,
The calculation unit calculates the first index value based on the first term stored in the first storage unit, and based on the second term stored in the second storage unit, the first term 2 Calculate the index value,
The evaluation apparatus according to claim 1. The calculation unit indicates a positive content or a negative content for each of the data collected from the information medium for each of the first index value and the second index value. To determine whether
The evaluation apparatus according to claim 1 or 2. The calculation unit weights the data including the first term and the data including the second term based on the type of the information medium, and calculates the first index value and the second index value. calculate,
The evaluation apparatus according to any one of claims 1 to 3. A generator that generates information that enables display of an image that compares the first index value and the second index value;
The evaluation apparatus according to any one of claims 1 to 4. From data collected from information media, extract data related to brands preset by companies or organizations,
In the extracted data, a first index value indicating the appearance frequency of the first term indicating the content expected by the company or the organization, and a second index value indicating the appearance frequency of the second term indicating the reaction of the customer, To calculate,
Evaluation method. On the computer,
From the data collected from information media, let us extract data related to brands set in advance by companies or organizations,
In the extracted data, a first index value indicating the appearance frequency of the first term indicating the content expected by the company or the organization, and a second index value indicating the appearance frequency of the second term indicating the reaction of the customer, To calculate
Evaluation program. An extractor for extracting data related to a brand preset by a company or organization from data collected from an information medium;
In the data extracted by the extraction unit, a first index value indicating the appearance frequency of the first term indicating the content expected by the company or the organization, and a second value indicating the appearance frequency of the second term indicating the reaction of the customer. An evaluation apparatus comprising: an index value; and a calculation unit that calculates an appearance frequency of a third term indicating the risk of the company or group and a third index value based on a type of the information medium. A first storage unit storing the first term;
A second storage unit storing the second term;
A third storage unit storing the third term, and
The calculation unit calculates the first index value based on the first term stored in the first storage unit, and based on the second term stored in the second storage unit, the first term 2 index value is calculated, and the third index value is calculated based on the third term stored in the third storage unit,
The evaluation apparatus according to claim 8. Whether the calculation unit indicates positive contents for each of the first index value, the second index value, and the third index value for each of the data collected from the information medium, or is negative To determine whether the content is correct
The evaluation apparatus according to claim 8 or 9. The calculation unit weights the data including the first term and the data including the second term based on the type of the information medium, and calculates the first index value and the second index value. calculate,
The evaluation device according to any one of claims 8 to 10. The calculation unit calculates the appearance frequency of the third term for each information medium, and based on the calculated appearance frequency of the third term for each information medium and a threshold set for each information medium. Calculating the third index value;
The evaluation device according to any one of claims 8 to 11. A generator that generates information that enables display of an image that compares the first index value, the second index value, and the third index value;
The evaluation device according to any one of claims 8 to 12. From data collected from information media, extract data related to brands preset by companies or organizations,
In the extracted data, a first index value indicating an appearance frequency of a first term indicating contents expected by the company or an organization, and a second index value indicating an appearance frequency of a second term indicating a consumer reaction Calculating a frequency of appearance of a third term indicating a risk of the company or group and a third index value based on a type of the information medium;
Evaluation method. On the computer,
From the data collected from information media, let us extract data related to brands set in advance by companies or organizations,
In the extracted data, a first index value indicating an appearance frequency of a first term indicating contents expected by the company or an organization, and a second index value indicating an appearance frequency of a second term indicating a consumer reaction , Calculating the appearance frequency of a third term indicating the risk of the company or group and a third index value based on the type of the information medium,
Evaluation program.

Images