JP2007241981A - Matching system and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a matching system and method capable of discriminating how far profile and tastes of a contributor match to those of browsers. <P>SOLUTION: Contributor information contributed from a client terminal (contributor terminal) 12 is compared with browser information input from a client terminal (browser terminal) 12 for browsing the contributor information and a degree of matching is calculated according to matching information. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a matching system and a matching method for calculating a degree of matching between poster information and viewer information.

Conventionally, for example, a site for posting word-of-mouth information on a specific product on a bulletin board on the Internet and browsing the word-of-mouth information has been operated (see Non-Patent Document 1).
[Search on February 1, 2006], Internet <URL: http://www.kakaku.com/index.html>

  However, the above-mentioned technology does not know whether the person who posted the word-of-mouth information is the same age group as the viewer or the owner of the same taste, and a guideline for how much the posted information is helpful to me There is no.

  The present invention has been made in consideration of the above-described circumstances, and an object thereof is to provide a matching system and a matching method capable of determining how much a poster's profile and preferences match a viewer.

  In order to achieve the above object, according to one aspect of the present invention, poster information posted from a poster terminal is compared with viewer information input from a browser terminal that browses the poster information. , A matching system for performing matching, and when there is access from the contributor terminal, the contributor terminal receives and stores the contributor information of the user of the contributor terminal, and access from the viewer terminal If there exists, the 2nd memory | storage means which receives and memorize | stores the viewer information of the user of the said browser terminal, the poster information memorize | stored in the said 1st memory | storage means, and memorize | stored in the said 2nd memory | storage means A matching system is provided, comprising: a calculating unit that compares the viewer information with each other and calculates the matching degree according to the degree of similarity.

  ADVANTAGE OF THE INVENTION According to this invention, it is possible to provide the matching system and the matching method which can discriminate | determine how much a submitter's profile and preference correspond with a reader.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 shows a block diagram of a matching system according to the first embodiment of the present invention. This matching system is realized as a server device 10 that provides information connected to a network 11 such as the Internet and each client terminal 12 connected to the network 11.

  As shown in FIG. 2, the server device 10 includes a control unit 20 that performs data matching processing and the like, a storage unit 21 that stores received data and processed data, and a communication unit 22 that connects to the network 11. ing.

  The storage means 21 includes a contributor information storage area 21a, a viewer information storage area 21b, and a matching information storage area 21c. The poster information storage area 21a is profile information (for example, age, sex) of a poster who posts information, interest information (for example, travel, outdoor), preference information (for example, a person who thinks about a car, usage pattern), etc. Remember the author's information. The browser information storage area 21b stores browser information such as profile information, interest information, and preference information of the viewer who browses the posted information. The matching information storage area 21c is rule information that determines how many matching points are given to each item (for example, age and gender of profile information) of the poster information and the viewer information. Remember.

  Next, FIG. 3 is a flowchart showing a matching method to which the matching system of the present invention is applied.

  The server device 10 stores (registers) the profile information in the poster information storage area 21a or the viewer information storage area 21b by access from the client terminal 12 (step S1).

  Note that there are two cases of access from the client terminal 12: a contributor or a viewer. Registration of the profile information of the poster or the viewer is displayed on the client terminal 12 as shown in FIG. 4, for example. The basic profile information of the poster or the viewer includes gender, age, unmarried, living environment (city or suburb), and the like. These are all required inputs (pressing a check button), for example. Further, the interest information includes travel (domestic travel, overseas travel, day trip, hot spring, drive, theme park), outdoor (fishing, auto camping, hiking / climbing, camping), and the like. These are, for example, one or more inputs. When the above input is completed, pressing the “Register” button completes the input of the basic profile information and interest information of the contributor or viewer.

  Next, answers to the questionnaire will be described. The server device 10 stores (registers) the preference information in the poster information storage area 21a or the viewer information storage area 21b by accessing from the client terminal 12 (questionnaire answer: step S2).

  For example, it is displayed on the client terminal 12 as shown in FIG. Information about the preferences of posters or viewers includes thoughts about cars (relaxing space, simple means of transportation, relieving stress like your own room, part of your life), usage (driving highways, driving urban areas, Commuting / commuting, work, daily shopping, family pick-up, weekend trip / vacation). For example, one or more items are selected from each item and required input. When the above input is completed, pressing the “Register” button completes the input of the preference information of the poster or the viewer.

  Furthermore, the server device 10 stores (registers) current affair information in the poster information storage area 21a or the viewer information storage area 21b by access from the client terminal 12.

  For example, it is displayed on the client terminal 12 as shown in FIG. As current information of a poster or a viewer, for example, Q. In response to the question of the postal privatization bill, Set three answer options: agree or disagree. Q. What is the new album of popular music group A? In response to the question “A. Three answer options are set: interested, not interested, or neither. When the above input is completed, pressing the “Register” button completes the input of current information of the poster or the viewer.

  For example, when the above input is completed, the server device 10 reads each piece of information stored in the poster information storage area 21a, the viewer information storage area 21b, and the matching information storage area 21c, and performs matching calculation (step S3). . For example, the matching information storage area 21c stores matching information such as 100 points if the gender matches among the items of the basic profile, and 0 points if the gender does not match, and the matching calculation is performed according to the matching information. The matching calculation is performed as follows, for example.

  Matching information is, for example, 100 for the same surname, 50 for the opposite sex, 100 for the age difference of 5 or less, 75 for the age difference of 6 to 10, and 50 for the age difference of 11 or more. In the case of unmarried, the match is 100 points and the mismatch is 50 points, and for other items, the option match is 100 points and the mismatch is 50 points.

  As shown in FIG. 7, the profile information of the viewer is gender: male, age 25, unmarried: unmarried, living environment: suburb, and the contributor is gender: male, age 32, unmarried: Marriage and living environment: If you are in the suburbs, your gender is 100, so the age is 75 because the age difference is 7 years old, unmarried is 50, and other items (or As an option), the living environment is the same in the suburbs, so it is 100 points, and the total matching degree of each item is 325 points. Then, the average of these items, which are gender, age, unmarried, and living environment, is 325/4 = 81.3 points (for example, rounded to the first decimal place).

  Further, when a questionnaire is taken with respect to two items of travel and outdoor as interest information, as shown in FIG. 8, the viewer selects domestic travel and overseas travel for the item “travel”, and item “ For “outdoors”, hiking and mountain climbing are selected. In addition, for the item “travel”, the contributor selects overseas travel and driving, and for the item “outdoor”, he selects fishing, hiking / climbing. In addition, as matching information, the matching options are 100 points, and in the case of non-matching, there are 50 points. Therefore, the matching items are overseas travel, day trip (non-selected match), hot spring, Theme parks, auto camps, hiking / climbing, and camping are 100 points each, and inconsistent domestic travel, driving and fishing are 50 points each. As a method for calculating the matching degree, for example, an average of the scores obtained by summing and averaging for each item is calculated. The total of the item “Travel” is 400 points, the average is 66.7 points (for example, rounded to the first decimal place), the total of the item “Outdoor” is 300 points, the average is 75 points, and the average of these two averages is , 70.9 points (for example, the second decimal place is rounded off). Further, all the options may be totaled to 700 points, and may be divided by 10 of the number of options to obtain an average of 70 points.

  Next, as a preference information, when a questionnaire is taken with respect to two items of an idea about a car and a usage form, as shown in FIG. Stress reduction is selected, and for the item “use form”, driving in the city area and daily shopping are selected. In addition, as matching information, the matching options are 100 points, and in the case of non-matching, there are 50 points. Therefore, the matching items are a relaxing space, simple moving means (non-selection matching), Like your own room (non-selected match), part of life (non-selected match), driving around the city, commuting to school (match non-selected), work (non-selected match), family pick-up (non-selected) 100 points each for selection, 50 points each for driving on the highway, shopping everyday, and weekend trips / vacations, like your room being inconsistent. As a method for calculating the matching degree, as described above, for example, an average of scores obtained by summing and averaging for each item is calculated. The total of the item “Thinking about the car” is 300 points, the average is 60 points, the total of the item “Usage form” is 400 points, and the average is 57.1 points (for example, rounded to the first decimal place). Further, the average is 58.6 points (for example, rounded to the first decimal place). Moreover, all the options may be totaled to 700 points, and may be divided by 12 as the number of options to obtain an average of 58.3 points.

  Next, as current affairs information, when a questionnaire is taken on two items of car thinking and usage, as shown in FIG. 9, the viewer agrees on the item “postal privatization” and item “ I am interested in "What is the new album of popular music group A?", The author is not interested in the item "What is the new album of popular music group A?" , Is selected. As matching information, the matching options are 100 points, and in the case of non-matching, the score is 50. Therefore, “Postal privatization”, which is a matching item, is 100 points and is a non-matching item. “What is the new album of popular music group A?” Is 50 points, and the average of the two items of 100 points is 50 points.

  Finally, when summing the above-mentioned matching degrees, a weight can be arbitrarily set for each item. For example, as shown in FIG. 11, the matching degree of the basic profile of the contributor and the viewer is 81.3, and this is multiplied by the weight A (for example, 10%) to match the interest information of the contributor and the viewer. Is 70.9, and this is multiplied by weight B (for example, 40%). The degree of matching between the poster and the viewer's preference information is 58.6, and this is multiplied by weight C (for example, 40%). The degree of matching between the current information and the viewer is 50, and this is multiplied by a weight D (for example, 10%). From the above, the total matching degree is 64.9 points.

  As described above, a viewer who has accessed an information site such as a word-of-mouth site has an interest, preference, etc. with the viewer (user) when browsing a comment on the product A of Mr. B who is the poster of the information. By displaying the similar point as the total matching degree, it is possible to refer to the contributor's opinion more similar to the viewer. Further, by the above-described matching degree calculation, the total matching degree with other contributors such as Mr. C, Mr. D, Mr. E, etc. can also be displayed.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to the drawings.

  FIG. 13 to FIG. 15 are schematic diagrams showing examples of question screens related to vehicles owned.

  For example, as shown in FIG. 15, in question 1, items such as “manufacturer”, “type”, “vehicle type”, “grade”, “year”, “use”, etc. You will be prompted to select one.

  In question 2, the user is prompted to select one or more reasons for purchasing the car he currently owns. At the end of Question 2, a pull-down menu is displayed so that the top three rankings can be selected for the selected answers.

  In question 3, it is displayed to select only one value for the car.

  In Q4, the items of “Manufacturer”, “Type”, and “Vehicle type” of the next car that the user wants to ride are displayed so as to select one from the pull-down menu.

  In Q5, you can select multiple types of cars you own at home, such as minicars, compact cars, sedan hatchbacks, wagons, minivans one box, sports coupes, SUVs (cross country), etc. A pull-down menu for selecting the number of units on the right side of each type is displayed. By answering these questions and pressing the “Next” button, the next question about the profile information is displayed.

  For questions regarding profile information, for example, as shown in FIG. Is displayed to select the gender.

  b. The date of birth is displayed in a pull-down menu. c. The selection as to whether or not you are married is displayed in a pull-down menu.

  d. As a selection of whether or not there is a child and a pull-down menu that can be selected when there is a child.

  e. As a pull-down menu for selecting the number of family members living together.

  f. The area where you live (for example, by prefecture) is displayed in a pull-down menu, and the location is a selection item of “more than 1 million cities”, “local core cities”, “suburbs other than cities”, or “other” Is displayed.

  Next, for example, as shown in FIG. In the case of a company employee, a pull-down menu for selecting a job type is displayed.

  h. Then, the household income is displayed in a pull-down menu for selecting from a predetermined range. i. Then select one or more that you are usually interested in.

  Next, as shown in FIG. Then, select one that you want to focus on in your future life. k. Then, select one favorite color. Subsequently, for example, as shown in FIG. Then, for all questions regarding self-confidence, choose one of four options: “I think so at all”, “I think so”, “I do n’t think so”, or “I do n’t think so” indicate.

  As described above, the matching method of the present invention will be described in the case where a question is asked and, for example, data of 200 or more men and women or a total of 400 or more is collected.

  A part of the data collected as described above is displayed as shown in FIG. 20, for example. FIG. 20 shows answers obtained about the motivation to purchase a car among the collected data. As mentioned above, multiple answers can be obtained regarding the motivation for purchasing a car, and the selected content can be grasped in the top 3 and 4th (not in the top 3). .

  For example, the number of people who chose “best in style” as the first motivation for purchasing a car is 16, of which 11 are men and 5 are women. In addition, the number of people who chose “best in style” as the second motivation for purchasing a car is seven, of which five are men and two are women. Furthermore, the number of people who chose “the style is good” as the third place as the motivation for purchasing a car is two, of which one is male and one is female. And the number of people who chose “4th style” or lower as the motivation for purchasing a car is two, of which one is male and one is female.

  In the same way, all the options such as “I have a distinctive image”, “I feel the quality of the appearance, I feel good” as the motivation for purchasing a car are ranked 1st to 3rd and 4th and below. Show the number of people selected and the number of men and women. And, for example, multiply the number of people selected by first place by 5 points, multiply the number of people chosen by second place by 3 points, and give 2 points to the number of people selected by third place. Multiply the number of people selected below 4th by 1 and calculate the total points 107. This total point is referred to as a weighted point. The weighted point of the item “style is good” is 107, of which 74 are weighted male points and 33 weighted female points. Similarly, weight points for each item are calculated.

  If the value obtained by adding all the weight points of each item is C, C is, for example, 2588. Next, the sum of the weighted male points and the weighted female points is multiplied by C and multiplied by 100 is referred to as weighted point weight (%). This weighted point weight is also calculated for each item. For example, in the item “style is good”, the weighted point weight is calculated as (74 + 33) / 2588 = about 4.1%. Of course, the sum of the weighted point weights for each item is 100%.

  Subsequently, if the absolute value of the difference between the weighted male points and the weighted female points is E, and the weighted point weight multiplied by E is F (correlation factor points), for example, the item “Style is good” Since E = 74−33 = 41, F = 41 * 4.1 (%) = 1.661 = about 1.7. Similarly, F (correlation factor point) is obtained for each item, and the total value of these Fs is 32.1.

  Similarly, correlation factor points are obtained for each profile factor (eg, sex, age, occupation, etc.), and the percentage of each correlation factor point is calculated.

  The calculated percentage is the weight used in the above-described first embodiment, and is used as the weight of the item “car purchase motive”. This weight is also called a matching profile weight.

  Thus, weighting is performed by calculating the matching profile weight for each item of the question and multiplying the result for each item. By performing such weighting, rational weighting reflecting the result of each item can be performed.

  Next, a case where there are three or more elements of the correlation factor point described above will be described with reference to FIGS.

  When there are three or more elements of correlation factor points, as shown in FIG. 21, for example, in the case of household annual income, the household annual income band is divided into six types of bands. In such a case, the top two elements with the highest ratio are selected. For example, “3 million yen to less than 5 million yen” and “5 million yen to less than 8 million yen” are the top two elements at 32.3% and 23.1%. Between these two elements, the absolute value of the difference between the above-described weighted point weight and the weighted point between the elements is calculated and multiplied to obtain a correlation factor point.

  In this way, even when there are three or more elements of the correlation factor point, the correlation factor point can be obtained as in the second embodiment. In addition, a correlation factor weight can be obtained for each profile factor by calculating a correlation factor point and calculating a percentage of each correlation factor point. For example, as shown in FIG. 22, a correlation factor point is obtained for each profile factor. Subsequently, as shown in FIG. 23, the percentage of each correlation factor point is calculated, and the matching profile weight for each profile factor is obtained.

  Next, a method for calculating cell matching points will be described in more detail. The cell matching points are indicated as ** points in the first embodiment described above.

  In the first embodiment, the cell matching points are calculated based on uniform matching information such as 100 points if the profile factors match and 0 points if they do not match. In this embodiment, cell matching points are calculated using the above-described weighted points. For example, when the weight point of one profile factor is 2/3 or more of the total points of the weight points, 0 points are set and 50 points are set when it is less than 2/3. For example, in the case of the item of “good style” as the above-mentioned car purchase motivation, the total points of the weighted points is 2588, two thirds thereof is 1725, and the weighted male points are 1586, two thirds. Since it is less than 50 points.

  Next, another example of the cell matching point calculation method in the case where there are three or more elements of the above-described correlation factor points will be described with reference to FIGS.

  When matching the above-mentioned household annual income data, since the household annual income band is divided into six types of bands, the calculation is performed as follows. In such a case, 100 points (for example, 100 points for “less than 3 million yen”) if the same bands coincide with each other, 20 points are subtracted every time one band is shifted. For example, if “less than 3 million yen” and “less than 3 million yen to less than 5 million yen”, one band shift is 80 points, and “less than 3 million yen” and “less than 5 million yen to less than 8 million yen” Because it is a band shift, it is 60 points, and if it is "less than 3 million yen" and "8 million yen to less than 10 million yen", it is 40 points because it is a shift of 3 bands, "less than 3 million yen" and "10 million yen to less than 15 million yen ”Is 20 points because of a shift of 4 bands, and“ less than 3 million yen ”and“ 15 million yen or more ”are 0 points because of a shift of 5 bands.

  Next, a cell matching point calculation method when the target to be matched is not between numerical parameters will be described with reference to FIG.

  As shown in FIG. 25, for example, 100 points are given for office workers (general occupations) that have the same occupation, 75 points for different occupations, and the like. Further, for example, if the company management is the same company management or a company officer, 100 points are assumed. As described above, regarding the matching method in the case where the object to be matched is not between numerical parameters, it is desirable to tentatively set cell matching points in light of social conventions and adjust them in the operation process.

  Similarly, as another example of the method of calculating the cell matching point when the target to be matched is not between numerical parameters, as shown in FIG. 26, between the site viewer and the contributor such as a comment. , 50 points for innovators, 50 points for innovators (viewers) and early adapters (contributors), 100 points for innovators (posters) and early adapters (viewers), etc. Can be set.

  Next, a processing method when a plurality of answers are obtained by multi-answer will be described.

  Multiply the number of matches between multiple answers obtained by multiple answers and multiply by the number of selectable items. For example, if a site visitor and a contributor have a “hobby” profile that can be selected from among 10 items, and matches 3 items of “Entertainment”, “Outdoors”, and “Car / Bike”, the following The cell matching points are calculated as follows.

  Number of matching items * 100 points / number of selectable items = 3 * 100/10 = 30 points. The matching profile weight described above is 30 points * 7.3% = 2.19 points for the item “hobby” of 7.3%, for example.

  As described above, by using the second embodiment, in addition to the same effects as those of the first embodiment, it is possible to further increase the matching accuracy.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. For example, do you include personality information (other, autonomous, cautious, aggressive, etc.) in your profile information, or possession information (cars, specific home appliances, real estate, financial assets, etc.)? Etc.). In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

The block diagram of the matching system which concerns on 1st Embodiment of this invention. The block diagram which showed the structure of the server apparatus which concerns on the matching system which concerns on 1st Embodiment of this invention. The flowchart which showed the matching method to which the matching system of this invention is applied. The schematic diagram which showed an example of the screen information of the selection screen displayed on a client terminal. The schematic diagram which showed an example of the screen information of the selection screen displayed on a client terminal. The schematic diagram which showed an example of the screen information of the selection screen displayed on a client terminal. The schematic diagram which showed an example of the matching method of basic profile information. The schematic diagram which showed an example of the matching method of interest information. The schematic diagram which showed an example of the matching method of preference information. The schematic diagram which showed an example of the matching method of current information. The schematic diagram which showed an example of the weight of the matching method. The schematic diagram which showed an example of the screen information on which the total matching degree was displayed for every contributor displayed on a client terminal. The schematic diagram which showed an example of the screen of the question regarding the car which has it concerning 2nd Embodiment of this invention. The schematic diagram which showed an example of the screen of the question regarding the car which has it concerning 2nd Embodiment of this invention. The schematic diagram which showed an example of the screen of the question regarding the car which has it concerning 2nd Embodiment of this invention. The schematic diagram which showed an example of the screen of the question regarding the profile which concerns on 2nd Embodiment of this invention. The schematic diagram which showed an example of the screen of the question regarding the profile which concerns on 2nd Embodiment of this invention. The schematic diagram which showed an example of the screen of the question regarding the profile which concerns on 2nd Embodiment of this invention. The schematic diagram which showed an example of the screen of the question regarding the profile which concerns on 2nd Embodiment of this invention. The schematic diagram which showed the data about the purchase motive of the car among the collected data. The schematic diagram which showed the ratio of the household annual income at the time of dividing into six bands. The schematic diagram which showed the correlation factor point with respect to each profile factor about the case where there are three or more elements of the correlation factor point. The schematic diagram which showed the matching / profile point with respect to each profile factor about the case where there are three or more elements of the correlation factor point. The schematic diagram which showed the calculation method of the cell matching point of household incomes when dividing into six zones | bands when there are three or more elements of a correlation factor point. The schematic diagram which showed the calculation method of the cell matching point when the object which matches is not numerical parameters. The schematic diagram which showed the calculation method of the cell matching point when the object which matches is not numerical parameters.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Server apparatus, 11 ... Network, 12 ... Client terminal, 20 ... Control means, 21 ... Storage means, 21a ... Contributor information storage area, 21b ... Viewer information storage area, 21c ... Matching information storage area, 22 ... Communication means.

Claims (12)

A matching system that performs matching by comparing poster information posted from a poster terminal with browser information input from a browser terminal that browses the poster information,
A first storage means for receiving and storing poster information of a user of the poster terminal when accessed from the poster terminal;
When there is access from the browser terminal, second storage means for receiving and storing viewer information of the user of the browser terminal;
A calculation unit that compares the poster information stored in the first storage unit with the viewer information stored in the second storage unit, and calculates a matching degree according to a similarity;
A matching system characterized by comprising: The matching system according to claim 1,
The said poster information is any one or more of the said author's profile information, interest information, preference information, and current affairs information, The matching system characterized by the above-mentioned. The matching system according to claim 1,
The matching system, wherein the viewer information is any one or more of the viewer's profile information, interest information, preference information, and current affairs information. In the matching system according to claim 1 or 3,
The calculation means compares predetermined items corresponding to the poster information and the viewer information, gives a predetermined score depending on whether or not they match, and calculates a matching degree from these scores. Matching system characterized by that. In the matching system according to claim 1 or 3,
The calculating means compares predetermined items corresponding to the poster information and the viewer information, assigns a predetermined score according to the degree of weighting, and calculates a matching degree from these scores. A matching system. The matching system according to claim 1,
A predetermined weighting is performed for each item on the predetermined score. The matching system according to claim 1,
When the number of the predetermined items to be compared is three or more, the matching system is characterized by comparing the top two selected items with the highest ratio. A matching method for performing matching by comparing the poster information posted from the poster terminal with the viewer information input from the browser terminal that browses the poster information,
A first storage step of receiving the poster information of the user of the poster terminal and storing it in a storage means when accessed from the poster terminal;
When there is access from the viewer terminal, a second storage step of receiving the viewer information of the user of the viewer terminal and storing it in the storage means;
A calculation step of comparing the poster information and the viewer information stored in the storage means, and calculating the matching degree according to the degree of similarity;
Matching method characterized by including The matching method according to claim 8,
The said poster information is any one or more among said poster's profile information, interest information, preference information, and current affairs information. The matching method according to claim 8,
The matching method, wherein the viewer information is any one or more of the viewer's profile information, interest information, preference information, and current affairs information. In the matching method according to claim 6 or 8,
The calculation means compares predetermined items corresponding to the poster information and the viewer information, gives a predetermined score depending on whether or not they match, and calculates a matching degree from these scores. A matching method characterized by that. The matching method according to claim 8,
A matching method, wherein predetermined weighting is performed for each item on the predetermined score.

Images