JP2003196301A - Keyword analysis system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a keyword analysis system by which information for supporting marketing is obtained by analyzing a keyword inputted by a consumer. <P>SOLUTION: The keyword inputted in a Web server 6 and the number of times of input of the keyword are acquired via a consumer systems 8a, 8b, 8c (a step S10) and moving average is calculated including the number of times of input of the day (a step S13). The latest moving average calculated in the step S13 is compared with moving average immediately before it, calculated based on the number of times of input for a prescribed period until the previous day, without including the day (a step S15) and when the latest moving average is increased for the moving average immediately before it by prescribed ratio or more (a step S16), the number of times of input of the day is compared with the number of times of input on the same day of the last week (a step S17). And the number of times of input on the day is increased for the number of times of input last week by prescribed ratio or more, the keyword is determined as a sharp advance keyword. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a keyword analysis system for obtaining information for marketing support by analyzing a keyword input by a consumer to a search engine site on a network.

[0002]

2. Description of the Related Art In recent years, with the spread of information communication equipment, consumers often use a network such as the Internet to obtain data about products and services. For example, by inputting a product name or the like on an Internet search engine site, detailed data regarding the product or the like is obtained.

As described above, since consumers use the Internet to obtain detailed data on products and the like, a large number of search keywords are input to search engine sites every day. The search keyword input by the consumer and the data on the number of times the keyword is input are valuable data that reflects the interest of the consumer, potential customer needs, and the like.

[0004]

As described above, since the search keyword and the input frequency data entered by the consumer are data reflecting the latest interests of the consumer, etc. By accurately analyzing the data input by, it is possible to obtain information about the product that is currently in the topic of consumers or the product that the consumer is paying attention to.

If the information indicating the interests of the consumer can be obtained, it is possible to predict the consumption trend of the consumer and to determine the so-called hot-selling product purchase timing and purchase amount. In addition, for example, in a store such as a convenience store where the life cycle of a product is short, a product or the like that is less interested in the consumer is specified based on the information indicating the consumption trend and the sale of such a product is terminated. It is possible to accurately determine the timing and the timing of product replacement. Therefore, an appropriate sales strategy can be planned based on such information indicating the interests of consumers.

An object of the present invention is to provide a keyword analysis system which obtains information for marketing support by analyzing a keyword input by a consumer.

[0007]

According to a first aspect of the present invention, there is provided a keyword analysis system, wherein the input data storage means stores an input keyword and the number of times the keyword is input in each first predetermined period, and the input data storage means. Calculated by the moving average calculation means for calculating a moving average of the number of input times of the keyword for each second predetermined period based on the number of input times for each of the first predetermined periods stored in Moving average storage means for storing the calculated moving average, moving average comparison means for comparing the latest moving average stored in the moving average storage means with a moving average immediately before the latest moving average, and the moving average As a result of comparison by the average comparing means, if the latest moving average is increased by a predetermined ratio or more with respect to the immediately preceding moving average, the keyword is jumped. Characterized in that it comprises a determining surge keyword determining means and the word.

According to the keyword analyzing system of the first aspect, for example, the first predetermined period such as every week is calculated based on the number of inputs every first predetermined period such as every day. It is determined whether or not the keyword is a soaring keyword based on the transition of the moving average of the number of inputs for each second predetermined period, which is a period that is a multiple of. That is, based on the transition of the moving average of the number of times of inputting the keyword, it is possible to accurately determine whether or not the keyword is a soaring keyword that is rapidly attracting high interest from consumers. Therefore, it is possible to obtain effective information for marketing support by accurately determining the soaring keyword.

Further, in the keyword analysis system according to claim 2, the moving average calculating means includes a moving average correcting means for correcting the moving average on the basis of the number of weekdays and the number of holidays within the second predetermined period. It is characterized by further comprising.

According to the keyword analyzing system of the second aspect, since the moving average is corrected based on the number of days of weekdays and the number of days of holidays, the influence of many holidays can be eliminated from the moving average. Therefore, it is possible to prevent the moving average from fluctuating because there are many holidays, to accurately determine the rapidly rising keyword, and obtain effective information for marketing support.

According to a third aspect of the present invention, there is provided the keyword analysis system, wherein the input data storage means stores the input keyword and the number of times the keyword is input in each first predetermined period, and the input data storage means stores the input data storage means. Among the number of input times for each of the first predetermined time periods, the average of the number of input times for each of the first predetermined time periods within the latest second predetermined time period that does not include the latest first predetermined time period is the average number of input times. An average number of input times calculation means, an input number comparison means for comparing the average number of input times with the latest number of input times of the first predetermined period stored in the input data storage means, and the number of input times comparison As a result of comparison by means, if the number of inputs in the latest first predetermined period increases by a predetermined ratio or more with respect to the average number of inputs, the keyword jumps. Characterized in that it comprises a determining surge keyword determining means and keyword.

According to the keyword analysis system of the third aspect, for example, the latest first number of inputs such as today's
By comparing the number of inputs in the predetermined period of, for example, with the average of the number of inputs every day in the second predetermined period that is a period that is a multiple of the first predetermined period such as yesterday to the previous week, It is determined whether the keyword is a skyrocketing keyword. Therefore, it is possible to accurately determine whether or not the keyword attracts high interest from consumers, and obtain information for planning an appropriate sales strategy.

According to another aspect of the keyword analysis system of the present invention, the latest number of inputs in the first predetermined period stored in the input data storage means and the first predetermined period in the past same day of the week. A day-of-week input count comparing means for comparing the number of inputs with the input number of days of the week is further provided, and as a result of the comparison by the day-of-week input count comparing means, as a result of the comparison by the day-of-week input count comparing means, It is characterized in that the keyword is determined to be a soaring keyword when the number of input times in the first predetermined period input on the same day increases by a predetermined rate or more.

According to the keyword analysis system of the fourth aspect, the latest moving average is compared with the immediately preceding moving average, or the number of inputs on the day is compared with the average number of inputs for a predetermined period not including the day. In addition, for example, the number of inputs in the latest first predetermined period, such as the number of inputs for today,
When the number of inputs on the same day in the past has increased by a predetermined ratio or more, it is determined that the keyword is a soaring keyword. Therefore, it is possible to accurately determine the soaring keyword and obtain useful information for marketing support.

According to a fifth aspect of the present invention, there is provided the keyword analysis system, wherein the input data storage means stores the input keyword and the number of times the keyword is input in each first predetermined period, and the input data storage means stores the input data storage means. A peak value storing means for storing the maximum number of inputs as a peak value among the number of inputs in the first predetermined period, and the latest number of inputs in the first predetermined period stored in the input data storage means. And the peak count stored in the peak value storage means are compared with the input count comparison means and the input count comparison means, and as a result, the latest input count in the first predetermined period becomes the peak value. On the other hand, when the ratio is less than or equal to a predetermined ratio, it is provided with a degree-of-interest decreasing keyword determining means for determining the keyword as a degree-of-interest decreasing keyword. And butterflies.

According to the keyword analysis system of the fifth aspect, for example, the latest first number of inputs such as today's
When the number of inputs in the predetermined period is less than or equal to the predetermined ratio with respect to the peak value, the keyword is determined to be the keyword with decreasing interest degree. That is, by comparing the number of times of past input with the highest number, it is possible to accurately determine the keyword of decreasing interest degree that the consumer is not interested in.
Therefore, it is possible to specify a product or the like that is less interested in the consumer, and obtain accurate information for replacing the product or changing the stocking amount of the product.

According to a sixth aspect of the present invention, there is provided a keyword analysis system, wherein the input data storage means stores the input keyword and the number of times the keyword is input in each first predetermined period, and the input data storage means stores the input data storage means. A first total value calculating means for calculating a total value of the number of inputs for each of the first predetermined periods within a second predetermined period including the latest first predetermined period, as a first total value; The total value of the number of inputs for each of the first predetermined periods within the latest second predetermined period that does not include the latest first predetermined period stored in the input data storage means is calculated as the second total value. Second total value calculating means, the first total value and the second total value calculation means
When the first total value is less than or equal to a predetermined ratio with respect to the second total value as a result of comparison by the total value comparison means for comparing the total value of And a sudden-fall keyword determining means for determining a sudden-fall keyword.

According to the keyword analysis system of the sixth aspect, for example, the first total value, which is the total value of the number of times of input for the week, is the total value of the number of times of input for the last week, for example. When the ratio is less than or equal to a predetermined ratio with respect to the total value of 2, the keyword is determined to be a sudden drop. That is, since it is determined whether or not the keyword is a sudden drop keyword based on the change in the total number of input times within a predetermined period, it is possible to accurately determine a keyword for which the consumer's interest is rapidly diminished. You can Therefore, we will obtain information for formulating an appropriate sales strategy, such as identifying products that have rapidly diminished consumer interest, and determining whether or not their interest in products is transient. be able to.

According to a seventh aspect of the present invention, the keyword analysis system corrects the missing value when the number of inputs in the first predetermined period stored in the input data storage means is a missing value. It is characterized by further comprising a missing value correction means.

According to the keyword analysis system of the seventh aspect, since the missing value of the input count in the first predetermined period is corrected, it is possible to prevent the input count data from becoming discontinuous and The continuity of the data of the number of times can be maintained, and it can be accurately determined whether the keyword is a keyword that is rapidly rising, a keyword of decreasing interest or a keyword that is rapidly falling.

Further, in the keyword analysis system according to claim 8, the keyword and the number of times of inputting the keyword are data which are input to a server by a consumer through electronic means and stored in the server, It is characterized by further comprising keyword acquisition means for acquiring the keyword and the number of times of input of the keyword from the server for each of the first predetermined periods.

According to the keyword analysis system of the eighth aspect, the keyword input by the consumer to the server and the number of input times of the keyword are acquired for each predetermined period, and the keyword jumps, the keyword decreases in interest, or the keyword drops. It is determined whether or not it is a keyword. Therefore, by accurately analyzing a large amount of data that accurately reflects consumers' interests, needs, etc., it is possible to accurately determine whether a keyword is a soaring keyword, a decreasing interest degree keyword, or a rapidly falling keyword. it can.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION A keyword analysis system according to an embodiment of the present invention will be described below with reference to the drawings. In the following, a case will be described as an example in which a search engine site established on the Internet performs analysis using a search keyword input by a consumer as analysis target data.

FIG. 1 is a block diagram of the entire system including the keyword analysis system according to the embodiment. The keyword analysis system 2 is connected via a network 4 to a Web server 6 of a company that has a search engine site on the network, and the Web server 6 consumes the consumption of each consumer via the network 4. System (personal computer, mobile terminal,
It is connected to terminals 8a, 8b and 8c such as mobile phones that can access search engine sites on the network.

The keyword analysis system 2 is a keyword in which the input keyword rapidly attracts the interest of the consumer (a steep keyword), a keyword in which the interest of the consumer has diminished (a keyword of decreasing interest), or a keyword of the consumer. The data processing unit 20 that performs processing such as determining whether or not the keyword is a keyword that is rapidly diminishing interest (falling keyword).
Is equipped with. The data processing unit 20 includes a keyword (search keyword) acquired from the Web server 6,
An input data storage unit 22 for storing the number of times of inputting the keyword and the moving average is connected. Further, in the data processing unit 20, when the communication control unit 24 that controls communication with the Web server 6 and the keyword are determined to be the keyword of the skyrocketing keyword, the keyword of decreasing interest degree, or the keyword of the sharp drop, the keyword is the keyword of the skyrocketing keyword, A display unit 26 is connected to display as a degree-of-interest decreasing keyword or a sudden-fall keyword.

Here, the input data stored in the input data storage unit 22 is data acquired from the Web server 6 via the communication control unit 24 every predetermined period, for example, every day, and each of the keywords is The number of inputs is stored in association with each keyword. For example, as shown in FIG. 2, “amino acid” on December 1, 2001
"15,430", "Vitamin""14,320","Negativeion""9,500","Polyphenol""1"
3,850 "and" Beer yeast "input times" 8,32 "
0 "or the like.

Next, referring to the flowchart of FIG.
A process of determining whether or not the input keyword is a soaring keyword will be described.

[0028] The consumer has the consumer system 8a, 8b, 8
When the search engine site is accessed via the network 4 using c and a search keyword is input, the input search keyword is stored in the Web server 6. That is, when the search keyword input by the consumer is input once, it is counted as one input count, and the keyword and the input count of each keyword are stored in the Web server 6 as input data.

In the keyword analysis system 2, W
The input data stored in the eb server 6 is acquired by the data processing unit 20 via the communication control unit 24 (step S10). That is, among the input data stored in the Web server 6, the top 10,000 input data in the order of the number of times of input per day are input from the Web server 6 to the keyword analysis system 2 once a day. The keyword and the number of input times of each keyword are acquired by the data processing unit 20 by being transmitted. The acquired keywords and the number of times each keyword is input are stored in the input data storage unit 22 as input data (see FIG. 2).

Next, it is confirmed whether or not there is a missing value in the input data (step S11). That is, the Web
The input data transmitted from the server 6 is not all the keywords input by the consumer and the number of input times, but
The top 10,000 keywords and the number of inputs of each keyword are listed in descending order of the number of inputs on that day. Therefore, there may be a keyword that is stored in the input data storage unit 22 but is not included in the keywords acquired on that day. For such a keyword, the number of inputs on that day becomes "0", and a missing value occurs. Therefore, the keyword acquired on that day is compared with the keyword stored in the input data storage unit 22, and it is confirmed whether or not there is a keyword having a missing value.

If the input data has a missing value (step S11), the missing value is corrected (step S12). In other words, even for keywords with missing values, there was a certain number of input until the previous day,
It is presumed that a certain number of times was input on that day as well. Therefore, half the number of input times of the keyword having the lowest number of input times in the input data acquired on that day is set as the number of input times on that day. If there is no keyword having a missing value (step S11),
Since it is not necessary to correct the missing value, the process directly proceeds to step S13.

Next, a moving average of the number of inputs including the number of inputs on that day is calculated (step S13), and the calculated moving average is stored (step S14). That is, for example, the day is December 1st, and when calculating the moving average for "brewer's yeast", one week including December 1st, that is, 1st from November 25th to December 1st A moving average (latest moving average) is calculated based on the number of inputs each day and stored in the input data storage unit 22 (see FIG. 2).

Here, the moving average is calculated in units of one week in order to eliminate the influence of the characteristics of the day of the week on the number of inputs. That is, as shown in the input count graph 32 displayed by black circles and solid lines in FIG. 4, the number of input keywords often differs significantly between weekdays and holidays. Therefore, if the number of input times of the keyword changes as shown in the input count graph 32, the moving average calculated on a weekly basis is as shown in the moving average graph 34 displayed by white circles and broken lines in FIG. The effect on the number of inputs due to the characteristics of the day of the week is eliminated and the value changes within a certain range.

Next, the latest moving average calculated including the number of inputs on that day and the immediately preceding moving average not including that day are compared (step S15). That is, step S
The latest moving average and the immediately preceding moving average calculated in 14, for example, the moving average from November 25 to December 1;
Compare with the moving average from November 24th to 30th.

Next, it is determined whether or not the moving average deviates by a predetermined ratio or more, that is, whether the latest moving average has increased by a predetermined ratio or more with respect to the immediately preceding moving average (step S1).
6). For example, the moving average from November 25th to December 1st (latest moving average) is a predetermined ratio or more with respect to the moving average from November 24th to 30th (moving average immediately before), for example, 5%. It is determined whether or not it is larger than the above.

If the latest moving average has increased by a predetermined ratio or more with respect to the immediately preceding moving average (step S1
6) The number of inputs on that day is compared with the number of past inputs on the same day of the week (step S17). That is, the number of inputs on that day and the number of inputs on the same day one week ago, for example,
The number of inputs on Saturday, December 1 and the number of inputs on Saturday, November 24 for "brewer's yeast" are compared. On the other hand, if the latest moving average has not increased by a predetermined ratio or more with respect to the immediately preceding moving average (step S1
6) The processing ends because it is not the keyword that has soared.

Next, the input count of the day and the input count of the same day in the past deviate from each other by a predetermined ratio or more, that is, the input count of the day increases by a predetermined ratio or more with respect to the input count of the same day in the past. It is determined whether or not (step S1)
8). For example, the number of inputs on Saturday, December 1 is 1
It is determined whether or not the number of inputs on Saturday, January 24th has increased by a predetermined ratio or more, for example, 25% or more.

When the number of inputs on that day has increased by a predetermined ratio or more with respect to the number of inputs on the same day in the past (step S18), it is determined that the keyword is a soaring keyword (step S19). That is, when the latest moving average has increased by a predetermined rate or more with respect to the immediately preceding moving average, and the number of inputs on that day has increased by a predetermined rate or more with respect to the number of inputs on the same day in the past. , That keyword is determined to be a skyrocketing keyword. Then, the keyword is displayed on the display unit 26 as a soaring keyword. On the other hand, if the number of inputs on that day has not increased by a predetermined ratio or more with respect to the number of inputs on the same day in the past (step S18), the processing ends because it is not a skyrocketing keyword.

Here, when it is determined that the keyword is a steep rise keyword, the degree of steep rise of the keyword may be indicated. That is, the degree of steep rise is calculated based on the increase rate of the latest moving average with respect to the immediately preceding moving average and the rate of increase with respect to the number of inputs on the same day in the past, and the degree of steep rise is displayed on the display unit 26 together with the steep rise keyword. Good.

The processing of steps S13 to S19 is performed for all the keywords included in the input data acquired in step S10.
If the keyword is determined to be a skyrocketing keyword in 19, the keyword is displayed on the display unit 26 as a skyrocketing keyword. Further, the keywords determined to be the skyrocketing keyword on that day may be listed as the skyrocketing keyword on that day.

According to this keyword analysis system, when the latest moving average has increased by a predetermined ratio or more with respect to the immediately preceding moving average, and the number of inputs on that day is the number of inputs on the same day in the past. On the other hand, it is determined that the keyword is a skyrocketing keyword when the number increases by a predetermined rate or more.
That is, it is possible to accurately determine whether or not the keyword is a surge keyword based on the variation in the number of inputs. Therefore, it is possible to obtain information for planning an appropriate sales strategy, such as accurately determining the sales start time, the sales period, the purchase time, the purchase amount, etc. of the product or the like.

If the obtained input data has a missing value, the missing value is corrected, and then it is determined whether or not the keyword is a soaring keyword. Therefore, it is possible to prevent the data of the moving average from becoming discontinuous, and to accurately determine whether or not the keyword is a soaring keyword while maintaining the continuity of the data.

Further, since it is determined whether the keyword is a skyrocketing keyword or not based on the data input by the consumer via a network such as the Internet, the consumer's input is shortly followed by the consumer's input. It is possible to understand the matters that are attracting attention.

In the above keyword analysis system, the latest moving average is 5% of the immediately preceding moving average.
The keyword that has increased more than the above and the number of times of input on that day has increased by 25% or more with respect to the number of times of input on the same day in the past is determined as a skyrocketing keyword. For example, if the latest moving average has increased by 10% or more from the immediately preceding moving average, and the number of inputs on that day has increased by 30% or more with respect to the number of inputs on the same day in the past, etc. May be determined. Also, a different increase rate may be set for each keyword.

Further, in the above-mentioned keyword analysis system, the moving average is calculated on a weekly basis without considering holidays and holidays. However, when there are holidays and holidays in one week, the moving average is calculated. The average may be corrected. In other words, if there are 4 weekdays and 3 holidays in a week, the number of times a keyword is input decreases on holidays, so moving averages are reduced compared to a normal weekday of 5 weekdays and 2 holidays. Becomes smaller. In such a case, the moving average may be calculated with 5 days on weekdays and 2 days on holidays. For example, as shown in FIG.
Although October 8th is Monday, it is a physical education day and is a holiday, so the number of inputs is significantly smaller than on other weekdays. Therefore, when calculating the moving average, the number of inputs for the 4 days from October 2 to 5 is multiplied by 5/4, and the number of inputs for the 6th to 8th is multiplied by 2/3. The moving average may be calculated based on the number of times.

Further, in the above-described keyword analysis system, it is judged whether or not the keyword is a surge keyword based on the transition of the moving average of the number of inputs and the comparison with the number of inputs on the same day in the past. It may be possible to determine whether or not the keyword is a soaring keyword based only on the transition of the moving average of the number of times. Even if the latest moving average is increased by a predetermined rate or more with respect to the immediately preceding moving average, even if the keyword is determined to be a skyrocketing keyword, the characteristics of the day of the week are excluded and the number of input times is moved. Since the average is calculated, it is possible to accurately determine whether or not the keyword is a soaring keyword.

Next, with reference to the flow chart shown in FIG. 5, another process for determining whether or not the input keyword is a soaring keyword will be described.

First, the Web server 6 to the communication control unit 24
The input data is acquired via (step S20), and it is confirmed whether or not a missing value occurs in the acquired input data (step S21). If a missing value has occurred, the missing value is corrected (step S22). on the other hand,
If no missing value has occurred, the process proceeds to step S23.
Since these processes correspond to steps S10 to S12 shown in FIG. 3, description thereof will be omitted.

Next, the average of the number of inputs in a predetermined period is calculated (step S23). That is, if the keyword “brewer's yeast” and the number of times of input are acquired for one week before the previous day stored in the input data storage unit 22 that does not include the day, for example, December 1st is not included. November 2
The average number of inputs is calculated based on the number of inputs every 4 to 30 days.

Next, the number of inputs on that day and step S23
It is compared with the average number of input times calculated in (step S24). For example, the number of inputs of "brewer's yeast" on December 1st is compared with the average number of inputs of "beer yeast" on November 24th to 30th.

Next, the average input count calculated in step S23 deviates from the input count of the day by a predetermined ratio or more, that is, whether the input count of the day increases by a predetermined ratio or more with respect to the average input count. It is determined whether or not (step S25). For example, it is determined whether or not the input count on December 1st has increased by a predetermined ratio or more, for example, 15% or more, with respect to the average input count on November 24th to 30th.

If the number of inputs on that day has increased by a predetermined ratio or more with respect to the average number of inputs (step S2
5) Then, comparison is made with the number of past inputs on the same day as that day (step S26). That is, the number of inputs on that day is compared with the number of inputs on the same day one week ago, for example, the number of inputs on Saturday, December 1 and the number of inputs on Saturday, November 24 for "beer yeast". On the other hand, if the number of input times on that day has not increased by a predetermined ratio or more with respect to the average number of input times (step S25), the processing ends because it is not a keyword that has soared.

Next, the input count of the day and the input count of the same day in the past deviate from each other by a predetermined ratio or more, that is, the input count of the day increases by a predetermined ratio or more with respect to the input count of the same day in the past. It is determined whether or not (step S2
7). For example, the number of inputs on Saturday, December 1 is 1
It is determined whether or not the number of inputs on Saturday, January 24th has increased by a predetermined ratio or more, for example, 25% or more.

When the number of inputs on that day has increased by a predetermined ratio or more with respect to the number of inputs on the same day in the past (step S27), it is determined that the keyword is a soaring keyword (step S28). For example, December 1
The number of input of "Beer yeast" on Saturday, November 25th
It has increased by more than 15% compared to the average number of inputs on 30th, and is 25 against the number of inputs on Saturday, November 24th.
When the increase is at least%, “brewer's yeast” is determined to be a soaring keyword. When “beer yeast” is determined to be a soaring keyword, it is displayed on the display unit 26 as a soaring keyword. On the other hand, if the number of inputs on that day has not increased by a predetermined ratio or more with respect to the number of inputs on the same day in the past (step S27), the keyword is not a skyrocketing keyword, and the process ends.

The processing of steps S23 to S28 is performed for all the keywords included in the input data acquired in step S20, and
When it is determined that the keyword is a soaring keyword in 28, the keyword is displayed on the display unit 26 as a soaring keyword.

According to this keyword analysis system, the number of inputs on that day has increased by a predetermined ratio or more with respect to the average number of inputs on the previous day excluding the day, and the number of inputs on the same day in the past. Also, in the case where the number is increasing by a predetermined ratio or more, it is determined that the keyword is a soaring keyword. Therefore, it is possible to obtain information for marketing support such as identifying a product or the like for which consumer interest is increasing and determining the purchase amount of such product by accurately determining the rapidly rising keywords. .

If the obtained input data has a missing value, the missing value is corrected, and then it is determined whether or not the keyword is a soaring keyword. Therefore, the continuity of the data can be maintained when calculating the average of the number of input times for one week up to the day before the input data is acquired, and it is more accurately determined whether the keyword is a surge keyword. be able to.

Since it is determined whether or not the keyword is a soaring keyword based on the number of times the keyword is input every day,
It is possible to accurately determine the time when the keyword starts to be noticed. Therefore, it can be used as information for determining the degree of fashion of products or the like corresponding to the keyword, and effective information for marketing support can be obtained.

In the above keyword analysis system, the number of inputs on the day is increased by 15% or more with respect to the average number of inputs for the week not including the day, and the input on the same day in the past. The keyword increasing by 25% or more with respect to the number of times is determined to be a sharply rising keyword, but the increasing rate can be appropriately selected. For example, when the increase rate with respect to the average number of input times is 25% or more and the increase rate with respect to the number of input times on the same day in the past is 30% or more, the keyword may be determined as a skyrocketing keyword.

Further, in the above-mentioned keyword analysis system, it is determined whether or not the keyword is a skyrocketing keyword based on the comparison with the average number of input times in a week up to the previous day not including the day and the number of input times on the same day in the past. However, it may be determined whether or not the keyword is a soaring keyword based only on the comparison with the average input count. When the number of input times on the day has increased by a predetermined ratio or more with respect to the average input number, even if the keyword is determined to be a skyrocketing keyword, by comparing with the average input number excluding the characteristics of the day of the week Since it is determined whether or not the keyword is the soaring keyword, it is possible to accurately determine whether or not the keyword is the soaring keyword.

Next, with reference to the flow chart shown in FIG. 6, a process of determining whether or not the input keyword is a keyword of decreasing interest degree will be described.

First, from the Web server 6 to the communication control unit 24
The input data is acquired via (step S30). The input data acquisition process is performed in step S1 shown in FIG.
Since the processing is the same as 0, the description is omitted.

Next, the peak value of the number of inputs is compared with the number of inputs on that day (step S31). Here, the peak value is the maximum number of inputs among the number of inputs stored in the input data storage unit 22 for each keyword. For example, as shown in FIG.
The highest number of inputs from 1st to 3rd December is 1.
The number of inputs on February 2 is "12,035".
2,035 "is the peak value of"brewer's yeast ".

If the number of inputs on that day is larger than the peak value (step S32), the peak value is updated (step S33). For example, when the keyword “brewer's yeast” and its input count are acquired on December 9, and the input count on December 9 is larger than the peak value, the input count on December 9 is changed to “brewer's yeast”. The peak value is updated as the peak value, and the processing is ended because it is not the keyword for decreasing the degree of interest.

On the other hand, when the number of inputs on the day is smaller than the peak value (step S32), it is determined whether the number of inputs on the day is less than or equal to a predetermined ratio with respect to the peak value (step S34). For example, it is determined whether or not the number of inputs of "brewer's yeast" on December 9 is 1/2 or less of the peak value.

The number of inputs on that day is 1 / the peak value
If the number is 2 or less (step S34), it is determined that the keyword is the interest level decreasing keyword (step S35). For example, 12
When the input count of “brewer's yeast” on the 9th of the month is 1/2 or less of the peak value, “brewer's yeast” is determined to be a keyword of decreasing interest level, and the display unit 26 displays the interest level. It is displayed as a reduced keyword. On the other hand, when the number of inputs on that day is not less than 1/2 of the peak value (step S34), it is not a keyword for decreasing the degree of interest, and the process ends.

The processing of steps S31 to S35 is performed for all the keywords contained in the input data acquired in step S30, and
When it is determined in 35 that the keyword is the interest level decreasing keyword, the keyword is displayed on the display unit 26 as the interest level decreasing keyword.

According to this keyword analysis system, the number of input times of the keyword on that day is 1/2 the peak value.
If the following is the case, it is determined that the keyword is of a degree of interest decrease. That is, it is possible to accurately determine whether or not the keyword has become less interested in consumers by comparing it with the peak value.

Further, since it is possible to accurately determine whether or not the keyword is a keyword with a decreased degree of interest, it is possible to accurately identify a product or the like for which the consumer's interest has diminished, and when to end the sale of such a product or the like. It can be used as a judgment material for determining the purchase quantity of products. Further, it can be used as information for planning an appropriate sales strategy in a retail store or the like, such as replacement of a product when the sale of a product or the like that has lost interest to consumers is finished.

In the above keyword analysis system, the case where no missing value exists in the input data has been described as an example. However, it is confirmed whether or not there is a missing value in the input data, and the missing value is confirmed. If there is a value, it may be determined whether or not the keyword is a keyword of decreasing interest degree after correcting the missing value. In other words, the number of times of input for the keyword with a missing value is “0” on that day, and since there is no data, it cannot be compared with the peak value, and whether the keyword has a degree of interest decrease or not. It becomes impossible to make a judgment.
Therefore, the missing value is corrected, that is, the number of times of inputting the keyword having the missing value is set to a value of 1/2 of the smallest number of times of input data acquired on that day. Then, by comparing the number of times of input with the peak value, it is determined whether or not the keyword is a degree-of-interest decreasing keyword.

Next, with reference to the flow chart shown in FIG. 7, another process for determining whether or not the input keyword is a sudden drop keyword will be described.

First, the Web server 6 to the communication control unit 24
The input data is acquired via (step S40). Note that the input data acquisition process is the same as step S10 of the flowchart shown in FIG.

Next, the total value of the number of inputs in a predetermined period is calculated based on the number of inputs per day stored in the input data storage unit 22 (step S41). Here, the total value is calculated on a weekly basis in order to eliminate the influence of the characteristics of the day of the week on the number of inputs. That is, the input data storage unit 2
2, for example, the number of input times for "brewer's yeast" per day is 1
If stored from February 1st to 21st, December 15th
The total value of the number of inputs per day from day 21 to 21 is calculated as the first total value, and the total value from December 8 to 14 is calculated as the second total value.

Next, the first total value calculated in step S41 and the second total value are compared (step S42). Then, it is determined whether or not the first total value is less than or equal to a predetermined ratio with respect to the second total value, for example, 1/2 or less (step S43).

The first total value is 1 / the second total value.
When the number is 2 or less (step S43), it is determined that the keyword is a sudden drop (step S44). Then, when it is determined that the keyword is a sudden drop keyword, it is displayed on the display unit 26 as a sudden drop keyword. On the other hand, if the first total value is not equal to or less than 1/2 of the second total value (step S4
3) The processing is ended because it is not a sudden drop keyword.

The processing of steps S41 to S44 is performed for all the keywords included in the input data acquired in step S40, and
When it is determined that the keyword is a sudden drop keyword in 44, the keyword is displayed on the display unit 26 as a sudden drop keyword.

According to this keyword analysis system, for example, the first total value, which is the total value of the number of input times in the week, is compared with the second total value, which is the total value of the input number in the last week. By doing so, it is determined whether or not the keyword is a sudden drop keyword. For example, when the input data storage unit 22 stores a keyword from December 1st to 21st and the number of input times of the keyword, a graph of the number of inputtings of the keyword "beer yeast" from December 1st to 21st Is shown in FIG. When the number of inputs is changing as shown in this figure, the total value 36 of this week is less than 1/2 of the total value 38 of the last week, and "brewer's yeast" is determined to be a sharp drop keyword. . Therefore, it is possible to accurately determine whether or not the input keyword is a suddenly falling keyword in which the interest of the consumer has suddenly decreased.

It is empirically known that the number of times a keyword is input is gradually decreased. However, by comparing the total number of input times within a predetermined period, the interest in that keyword can be determined. Based on the degree of fading, it is possible to identify a product or the like that consumers are no longer interested in, and obtain information for marketing support that accurately determines the time or the like for product replacement.

Further, for example, when a certain product is introduced on the television and the keyword related to the product becomes a sharply rising keyword, but when it turns into a sharply falling keyword within a short period of time, the consumer of the keyword is changed. It can be determined that the interest was transient. Therefore,
It is possible to obtain effective information for determining the amount of goods purchased.

In the above keyword analysis system, it is not confirmed whether or not there is a missing value in the input data, but it is confirmed whether or not there is a missing value in the input data. If there is a missing value, the missing value may be corrected, and then it may be determined whether the keyword is a sudden drop. That is, after correcting the missing value, 1
By calculating the total number of input times for the week and comparing the total value of the number of input times for the week with the total value of the number of input times for the previous week, it is possible to determine whether or not the keyword is a sudden drop. Good.

[0081]

According to the present invention, whether or not a keyword is a soaring keyword that is rapidly attracting high interest from consumers based on the transition of the moving average of the number of times of inputting the keyword or the variation of the number of times of inputting the keyword. It can be accurately determined. Therefore, by accurately determining the soaring keyword, it is possible to obtain information for planning an appropriate sales strategy.

Further, by comparing the number of times of past input with the largest number, it is possible to accurately determine whether or not the keyword is a keyword of decreasing interest degree. Therefore, it is possible to obtain effective information for marketing support, for example, by identifying a product or the like for which the consumer is less interested and determining the purchase amount of the product.

Further, it is possible to determine whether or not the keyword is a sudden drop keyword based on the variation in the total number of input times within a predetermined period. Therefore, it is possible to obtain effective information for marketing support such as identifying a product etc. where the consumer's interest has suddenly diminished and determining whether the interest in the product was transient or not. it can.

Further, since the missing value of the number of inputs in the first predetermined period is corrected, the data of the number of inputs is prevented from becoming discontinuous, and the continuity of the data of the number of inputs is maintained to keep the keyword. It is possible to accurately determine whether or not is a rapidly rising keyword, a keyword of decreasing interest degree, or a rapidly falling keyword.

Further, by accurately analyzing a large amount of data that accurately reflects the interests and needs of consumers, it is possible to accurately determine whether or not the keyword is a keyword that is rapidly rising, a keyword that is decreasing in interest level or a keyword that is rapidly falling. You can go and get useful marketing information.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of an entire system including a keyword analysis system according to an exemplary embodiment of the present invention.

FIG. 2 is a diagram showing input data according to the embodiment of the present invention.

FIG. 3 is a flowchart for explaining a process of determining whether or not the keyword is a soaring keyword in the keyword analysis system according to the embodiment of the present invention.

FIG. 4 is a diagram for explaining a moving average of the number of times a keyword is input according to the embodiment of the present invention.

FIG. 5 is a flowchart for explaining another process of determining whether or not the keyword is a soaring keyword in the keyword analysis system according to the embodiment of the present invention.

FIG. 6 is a flowchart for explaining a process of determining whether or not the keyword is a keyword of decreasing interest degree in the keyword analysis system according to the exemplary embodiment of the present invention.

FIG. 7 is a flowchart for explaining another process of determining whether or not a keyword is a sudden drop in the keyword analysis system according to the embodiment of the present invention.

FIG. 8 is a diagram showing an example of a total value of the number of times a keyword is input according to the embodiment of the present invention.

[Explanation of symbols]

2 ... Keyword analysis system, 4 ... Network, 6 ...
Web server, 8a, 8b, 8c ... Consumer system, 2
0 ... Data processing unit, 22 ... Input data storage unit, 24 ... Communication control unit, 26 ... Display unit, 32 ... Input frequency graph, 34
… Moving average graph, 36… Total value of this week, 38… Total value of last week.

Continued front page    (72) Inventor Eisuke Sudo             2-2-1 Otemachi, Chiyoda-ku, Tokyo Stock             Inside the Nomura Research Institute F term (reference) 5B075 NS10 UU40

Claims (8)

[Claims] 1. An input data storage unit for storing an input keyword and the number of times the keyword has been input for each first predetermined period, and an input for each first predetermined period stored in the input data storage unit. The second of the keywords based on the number of times
A moving average calculation means for calculating a moving average of the number of inputs for each predetermined period, a moving average storage means for storing the moving average calculated by the moving average calculation means, and a latest moving average stored in the moving average storage means. Moving average of the moving average immediately before the latest moving average, and the moving average comparing unit compares the moving average with the moving average immediately before the latest moving average. A keyword analysis system comprising: a soaring keyword determining means for determining the keyword as a soaring keyword when the ratio increases by a ratio or more. 2. The moving average calculation means further comprises moving average correction means for correcting the moving average based on the number of weekdays and the number of holidays within the second predetermined period. The described keyword analysis system. 3. An input data storage unit for storing the input keyword and the number of times the keyword has been input in each first predetermined period, and an input for each first predetermined period stored in the input data storage unit. Among the number of times, an average input number calculation means for calculating, as an average input number, an average of the input numbers for each of the first predetermined periods within the latest second predetermined period that does not include the latest first predetermined period, The latest input count comparing means for comparing the latest input count and the average input count stored in the input data storage means with the average input count, And a keyword jumping determination means for determining the keyword as a keyword jumping when the number of inputs in a predetermined period of 1 has increased by a predetermined rate or more with respect to the average number of inputs. A keyword analysis system comprising: 4. A day-of-week input count comparison for comparing the latest input count of the first predetermined period stored in the input data storage means with the input count of the past same first day of the first predetermined period. As a result of the comparison by the day-of-week input count comparing means, the sharpness keyword determining means further determines that the latest input count of the first predetermined period is within the first predetermined period of the same past day. The keyword analysis system according to any one of claims 1 to 3, wherein the keyword is determined to be a soaring keyword when the number of inputs is increased by a predetermined ratio or more. 5. An input data storage unit that stores the input keyword and the number of times the keyword has been input in each first predetermined period, and the number of inputs in the first predetermined period that is stored in the input data storage unit. Of these, the peak value storage means for storing the largest number of inputs as a peak value, the latest number of inputs in the first predetermined period stored in the input data storage means, and the peak value storage means When the number of inputs in the latest first predetermined period is less than or equal to a predetermined ratio with respect to the peak value as a result of comparison by the number of inputs comparison means for comparing the peak value with the input number comparison means. A keyword analysis system, comprising: a keyword for decreasing a degree of interest for determining the keyword as a keyword for decreasing a degree of interest. 6. An input data storage unit that stores the input keyword and the number of times the keyword has been input in each first predetermined period, and the latest first predetermined period stored in the input data storage unit. A first total value calculating means for calculating a total value of the number of inputs for each of the first predetermined periods within a second predetermined period including as a first total value; and a latest total value stored in the input data storage means. Second total value calculation means for calculating, as a second total value, a total value of the number of inputs for each of the first predetermined periods within the latest second predetermined period that does not include the first predetermined period of As a result of the comparison by the total value comparing means for comparing the first total value and the second total value with the total value comparing means, the first total value is predetermined with respect to the second total value. If the ratio is less than or equal to Keyword analysis system, characterized in that it comprises a determining plunge keyword determining means and plunged keywords. 7. A missing value correction means for correcting the missing value when the number of inputs in the first predetermined period stored in the input data storage means is a missing value. The keyword analysis system according to any one of claims 1 to 6. 8. The keyword and the number of times of inputting the keyword are data input to a server by a consumer through an electronic means and stored in the server, and the data are stored in the server for the first predetermined period. The keyword analysis system according to any one of claims 1 to 7, further comprising a keyword acquisition unit that acquires the keyword and the number of times of input of the keyword for each time.