KR102509553B1 - Analyzing comsumption pattern and predicting future demand using bigdata analysis framework platform - Google Patents

Abstract

The present invention discloses a consumption pattern analysis and future demand prediction system, which includes: a data collection unit which selects evaluation factors which affect product demand and collects data of the evaluation factors; a data analysis unit which recombines the evaluation factors by calculating the relationship and weight of the evaluation factors; and a demand forecasting unit which predicts the future demand of the product based on the data recombined in the data analysis unit.

Description

Consumption pattern analysis and future demand prediction system using big data analysis framework platform {ANALYZING COMSUMPTION PATTERN AND PREDICTING FUTURE DEMAND USING BIGDATA ANALYSIS FRAMEWORK PLATFORM}

The present invention relates to a system for analyzing consumption patterns and predicting future demand, and more particularly, by analyzing past consumption patterns using an online Bigdata Analysis Framework Platform (hereinafter referred to as 'BAFP') to predict future demand. It is about a 'consumption pattern analysis and future demand prediction system using a big data analysis framework platform' that can predict demand.

With the popularization of mobile devices, the use of web information is becoming more and more common, and many people freely express their opinions through online media such as bulletin boards of specific websites, blogs, community cafes, and social network services.

When users evaluate the value of a particular piece of information, they often make decisions based on the opinions of other users online.

For example, through the Internet or mobile applications, consumers express various opinions ranging from product reviews to restaurant reviews.

These opinions can be used as data for other consumers to refer to before purchasing the product or going to a restaurant, and from the point of view of the service provider, it can be used as big data when indirectly wanting to know the position of the consumer.

Recently, with the spread of the digital economy, the big data environment in which an immeasurable amount of information and data is produced around us is expanding.

Compared to the data produced in the analog environment in the past, big data is vast in size, has a short generation cycle, and includes large-scale data including text and image data as well as numerical data in form. analyzed and used.

However, even if such a vast amount of big data is used, it is not easy to predict the future demand of consumers for a specific product beyond identifying the past trends or purchasing tendencies of consumers.

Republic of Korea Patent Publication No. 10-2019-0132016 Republic of Korea Patent Publication No. 10-2020-0055835 Republic of Korea Patent No. 10-1784559 Republic of Korea Patent No. 10-1922031 Republic of Korea Patent No. 10-2347224

The present invention has been made to solve the above problems, and an object of the present invention is to provide a system capable of analyzing consumption patterns of consumers by combining the rate of change of various factors that are related to each other.

In addition, an object of the present invention is to provide a system capable of predicting future demand for a specific product based on the consumption patterns of consumers analyzed by combining various past data.

In addition, an object of the present invention is to provide a calculation algorithm capable of predicting future demand for a product by synthesizing data of various factors affecting demand for the product.

In order to solve the above problems, the present invention selects evaluation factors that affect product demand and collects data of evaluation factors, and a data collection unit that calculates the relationship and weight of the evaluation factors to recombine the evaluation factors. Disclosed a consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, comprising a data analysis unit and a demand forecasting unit for predicting future demand for products based on the data recombined in the data analysis unit. do.

According to an embodiment of the present invention, the data analysis unit calculates a relationship between the selected evaluation factors, groups the evaluation factors based on the calculated relationship, sets a rank of the grouped evaluation factors, and ranks the evaluation factors. Disclosed is a consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that a demand forecast graph is generated by merging data of evaluation elements with low rankings with high evaluation elements.

According to an embodiment of the present invention, the demand forecasting unit outputs demand forecast data of a product selected through the demand forecast graph, and compares demand forecast data of the selected product and similar products to obtain demand forecast data of the selected product. Disclosed is a consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that it is formed to correct.

According to an embodiment of the present invention, consumption pattern analysis using the big data analysis framework platform, characterized in that the rank of the evaluation factors has a higher rank as the difference between the maximum value and the minimum value of the data within the set interval range increases. and a future demand forecasting system.

According to an embodiment of the present invention, the data analysis unit multiplies data of other evaluation factors except for the evaluation element of the highest rank (first rank) by a conversion factor to match the range of change of the data with the first rank data, and then in the first section. Disclosed is a consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that data is corrected so that evaluation factors have the same value.

According to an embodiment of the present invention, the data analysis unit estimates a polynomial trend line of the evaluation factors, creates a differential graph of the polynomial trend line, identifies a location of an inflection point, and determines the location of an inflection point, and data of the evaluation factors based on the location of the inflection point. Disclosed is a system for analyzing consumption patterns and predicting future demand using a big data analysis framework platform, characterized in that a demand forecast graph is created by merging.

According to an embodiment of the present invention, the demand forecasting unit predicts future demand based on a ratio between a value of a first-rank evaluation factor predicted by the demand prediction graph and a value of a recently measured first-rank evaluation factor. consumption pattern analysis and future demand forecasting system using a big data analysis framework platform.

According to an embodiment of the present invention, in the data merging of the evaluation factors, the inflection point of the trend line of the evaluation elements of the second rank (second rank) is identified, and the data of the first rank evaluation elements are transferred to the second rank evaluation elements in the intervals after the inflection point. Disclosed is a consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that the data of the evaluation factor and the average value of the data of the first ranking evaluation element are substituted.

According to an embodiment of the present invention, the demand forecast graph is big data, characterized in that it is drawn as a polynomial trend line from the section immediately before the inflection point to the section after the recently measured section among the data of the merged first ranking factors. Introduce a consumption pattern analysis and future demand forecasting system using an analysis framework platform.

According to an embodiment of the present invention, the demand forecast graph is a consumption pattern analysis and future using a big data analysis framework platform, characterized in that the polynomial trend line of the degree obtained by adding 1 to the number of inflection points of the second-order evaluation factor. Initiate a demand forecasting system.

According to the present invention, it is possible to increase the utilization as a market analysis model by predicting market demand by analyzing data acquired in real time from the market in operation.

In addition, the big data analysis module (BAFP) presented in the present invention is designed to utilize a large amount of open and shared data in the big data era, enabling accurate analysis and increasing the target scalability of big data analysis. It makes it possible to analyze country-by-country.

In addition, according to the present invention, analysis by generation and country is possible according to consumption patterns, and customized demand prediction according to each target is possible using analysis of behavior types between generations and overseas consumption patterns.

In addition, in the present invention, it is possible to more accurately predict the change rate of demand by ranking a plurality of factors influencing demand and merging the rate of change of a low-ranking factor with the rate of change of a high-ranking factor.

1 is a conceptual diagram showing a data processing structure of a functional module of a big data analysis module (BAFP) according to an embodiment of the present invention.
2 is a diagram conceptually illustrating a structure of processing multiple tasks of a big data analysis module (BAFP) according to an embodiment of the present invention.
3 is a diagram for explaining an operating structure of a future demand forecasting system according to an embodiment of the present invention.
4 is a flowchart illustrating a method of analyzing consumption patterns and predicting future demand according to an embodiment of the present invention.
FIG. 5 is a flowchart for specifically explaining the 'step of recombining evaluation elements' shown in FIG. 4 .
FIG. 6 is a flowchart for explaining in detail the 'step of predicting demand for a product' shown in FIG. 4 .
FIG. 7 is a flowchart for explaining in detail the 'step of generating a demand forecasting graph' shown in FIG. 5 .
8 is a diagram showing examples of fields in which BAFP according to the present invention can be utilized.
9A is a diagram illustrating a step of generating an original graph in order to generate a demand forecasting graph.
9B is a diagram illustrating a step of generating an initial value match graph according to magnification conversion in order to generate a demand forecast graph.
9C is a diagram for explaining a step of generating a rise and initial value correction graph in order to generate a demand forecast graph.
9D is a diagram illustrating a step of generating a polynomial trend line graph in order to generate a demand forecast graph.
9E is a diagram illustrating a step of generating a differential graph of a polynomial trend line in order to generate a demand forecast graph.
9F is a diagram illustrating a step of generating a primary merged graph in order to generate a demand forecasting graph.
9G is a diagram explaining a step of generating a secondary merged graph in order to generate a demand forecasting graph.
9H is a diagram explaining a step of generating a tertiary merged graph in order to generate a demand forecasting graph.
9H is a diagram for explaining a method of predicting future demand through a trend prediction graph.
10 is a diagram showing an example of reporting based on future demand prediction results.

Hereinafter, the present invention will be described in more detail with reference to the drawings. In this specification, the same or similar reference numerals are assigned to the same or similar components even in different embodiments, and the description is replaced with the first description. Singular expressions used herein include plural expressions unless the context clearly dictates otherwise.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Terms used in this specification are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

In addition, terms such as "...unit", "...unit", and "...module" described in the specification mean a unit that processes at least one function or operation, which is hardware, software, or hardware and It can be implemented as a combination of software.

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the present invention, if it is determined that a detailed description of related known technologies may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

1 is a conceptual diagram showing a data processing structure of a functional module of a big data analysis module (BAFP) according to an embodiment of the present invention.

Referring to FIG. 1, the big data analysis module (BAFP) configures a calculation set by defining the correlation of data collected in real time, and calculates the calculated set data (case 1, case 2, case 3, etc.) Errors can be corrected by comparison.

As shown, the work performed by the big data analysis module (BAFP) can be largely divided into three steps.

In the first step (I), a set of evaluation factors is formed by defining the correlation of the data collected in real time.

For example, if you want to analyze online sales, you include factors related to sales in your set. In this case, the evaluation factor set may include the number of views, the number of likes, the number of comments, reviews, ratings, hash tags, and the like.

The big data analysis module (BAFP) can calculate the correlation between sales volume and evaluation factors, exclude elements with low correlation from the set, and add elements with high correlation to the set. For example, if it is analyzed that the number of likes is rapidly decreasing while the number of sales is increasing, the module may evaluate the correlation between the number of likes and sales as low and exclude the number of likes data from the set.

In the second step (II), the evaluation elements are recombined by calculating the relationship and weight of the evaluation elements included in the set.

For example, the BAFP module calculates the relationship and weight between the number of comments data (B), the rating average data (C), and the number of views data (D), and recombines evaluation elements with a high degree of relationship. Data can be recombined by merging evaluation factors with low weight.

When recombining, only two pieces of data may be recombined, or all related data may be recombined when it is determined that a plurality of pieces of data have a high degree of relationship.

In FIG. 1, case 1 represents recombination data of ABC, case 2 represents recombination data of ACD, and case 3 represents recombination data of BCD. In each case, errors may occur because data are extracted using different evaluation factors. BAFP can perform step III work after correcting errors by comparing data extracted from each case.

In the third step (III), the evaluation elements are recombined by calculating the relationship and weight of the evaluation elements included in the set.

In step III, the data values derived in step II are intersected and sequentially calculated using modules such as the Pattern Classification Module, A Correlation Coefficient Module, and Weighted Ratio Module. It can be processed with predictable data.

The pattern classification module, A correlation coefficient module, and weighted ratio module shown in FIG. 1 are merely examples for convenience of explanation, and these modules are necessarily included in the present invention. need not be included in

2 is a diagram conceptually illustrating a structure of processing multiple tasks of a big data analysis module (BAFP) according to an embodiment of the present invention.

In the present invention, each task may be data collected through different paths of equal levels. For example, when purchasing a product online, consumers can use domestic open markets (11st, Auction, G Market, etc.), overseas shopping malls (Amazon, Ali Express, etc.) Tail, etc.), etc. to purchase the same product.

In this way, even the same product can be purchased through different channels, and it is not easy to directly compare one-to-one because each site has a different way of purchasing or recording buyer feedback.

In the present invention, a plurality of tasks are provided to respond to various environments in collecting and introducing big data online. For each task, individual information collection can be performed on the third party website. Since the type of data collected and the collection environment are different for each information collection site, a module suitable for the environment can be configured as a task.

The big data analysis module can create a more accurate future demand forecasting mechanism by analyzing the correlation of the data collected through each task and extracting common elements.

The big data analysis module (BAFP) can classify different variables collected in different environments according to the purpose of analysis, store them in a log data database (Log Data DB), and perform comparative analysis. In addition, the big data analysis module (BAFP) can store data useful for analysis by processing comparatively analyzed data into a modular cloud data set.

The big data analysis module (BAFP) analyzes the correlation of elements (variables) that collect necessary information from each other company's homepage site, compares it with other extraction groups, and inversely calculates the data that could not be extracted to infer the data value. For example, after analyzing the sales volume, view, and review of Task 1 (T1) to Task N (TN), the sales volume and number of views can be inferred through the review of Task-i, and the big data analysis module (BAFP) is an element that collects necessary information from other companies' homepages, and it is possible to infer more accurate data values by applying weights such as industry, classification, and price.

3 is a diagram for explaining an operating structure of a future demand forecasting system according to an embodiment of the present invention.

Referring to FIG. 3 , the big data analysis module (BAFP) may be connected to a user, another server, and an Internet communication network through a server.

The big data analysis module (BAFP) may include a data collection unit, a data analysis unit, a demand forecasting unit, a management unit, and the like. These 'parts' are only classified to explain the function of the big data analysis module, and may not be separated in terms of hardware. In addition, it is possible that there are more 'units' other than the 'unit' shown in FIG. 3 or that a plurality of 'units' are combined into one.

The data collection unit may include a communication module 111 and a database 112 .

The communication module 111 may include one or more modules enabling wired/wireless communication between the big data analysis module and the server.

The database 112 stores data collected by the big data analysis module and data collected, analyzed, and processed by the data analysis unit 120, the demand forecasting unit 130, and the management unit 140.

The data collection unit 110 may collect data using a web crawler method.

A web crawler is a computer program that navigates the World Wide Web in an organized and automated way. In general, the work that web crawlers do is called web crawling or spidering. Many sites, such as search engines, perform web crawling to keep their data up to date. Web crawlers are used to create copies of every page on the site they visit, and search engines index these pages for quick retrieval. Crawlers are also used for automatic maintenance of web sites, such as link checking and HTML code verification, and are also used to collect certain types of information from web pages, such as automatic e-mail collection.

A web crawler is a form of bot or software agent. Web crawlers usually start with a list of URLs, called seeds, and update the URL list by recognizing every hyperlink on a page. The list of updated URLs is revisited recursively.

The operating structure of the computer program search engine, which has the ability to search the World Wide Web in an organized and automated way, operates in three stages: web crawling, indexing, and searching.

First, the web crawling process copies all the pages of your site. This is done with the latest data first. And, the indexing process collects, stores, and analyzes data to enable accurate and fast retrieval. Finally, the extraction process extracts titles and subjects that contain the words constituting the search term.

The search is performed through this series of processes. The retrieval ability of a search engine is determined by the relationship between a search term and data. Among the many searched web pages, there are pages with relatively higher relevance to the search keyword, and the search engine continuously provides feedback through indexing analysis of the most relevant results for priority values adjust to

The data analysis unit 120 analyzes the data collected by the data collection unit 110 and edits or processes the data into a form usable by users.

For example, the data analysis unit 120 selects evaluation factors that are correlated with each other within the big data, calculates the relationship and weight of the evaluation factors, groups the evaluation factors according to the relationship, or assigns weights to the evaluation factors. It is possible to process big data into a form that is meaningful to the user by performing tasks such as arranging according to the data.

The demand forecasting unit 130 generates a future demand forecasting mechanism based on the data analyzed by the data analysis unit 120 to predict the future demand of consumers for a specific product.

4 is a flow chart illustrating a consumption pattern analysis and future demand forecasting method according to an embodiment of the present invention, and FIGS. 5 to 7 are flowcharts illustrating this in more detail.

Specifically, in FIG. 5, the 'step of recombining evaluation factors' is described in detail, in FIG. 6, the 'step of predicting product demand' is specifically described, and in FIG. 7, the 'step of generating a demand prediction graph' is described in detail. Explain in detail.

Referring to FIG. 4, the consumption pattern analysis and future demand forecasting method includes selecting evaluation factors (S100), collecting data of evaluation factors (S200), and calculating the relationship between evaluation factors and their weights to evaluate evaluation factors. It includes a step of recombination (S300) and a step of predicting demand for a product (S400).

The order of selecting evaluation factors (S100) and collecting data of evaluation factors (S200) may be reversed. In other words, the system can collect data on the evaluation factors after selecting specific evaluation factors, or extract data on the evaluation factors by selecting evaluation factors from among already collected data.

Evaluation factors may include the number of sales, the number of views, the number of likes, the number of comments, the number or content of reviews, the rating, the number or content of hash tags, and the like.

Referring to FIG. 1, the step of selecting evaluation factors (S100) is performed in step I. Evaluation factors such as sales volume, number of views, number of likes, number of comments, number or content of reviews, rating, number or content of hash tag described above may correspond to data A to D displayed in step I.

In the step of collecting data of evaluation factors (S200), the system collects data through the data collection unit 110. As described above, the data collection unit 110 may generate big data by collecting data in a web crawler method and storing the data in a database.

In the step of recombining the evaluation factors by calculating the relationship and weight of the evaluation factors (S300), the data of the evaluation elements selected and collected in the steps S100 and S200 are analyzed and processed into a form that can be used for demand forecasting.

According to an embodiment of the present invention, the step of recombining evaluation factors (S300) includes calculating a relationship between evaluation factors (S310), grouping evaluation factors according to the relationship (S320), and evaluating within a group. The step of arranging elements according to their weights (S330), the step of displaying a graph of the rate of change of evaluation factors (S430), merging the graph of evaluation elements with high weight with the graph of evaluation elements with low weight to create a demand forecasting graph It may include a step of generating (S350) and the like.

In the step of calculating the relationship between the evaluation factors (S310), the tendency between the evaluation factors is analyzed. For example, if sales volume increases when the number of views or comments on a specific product increases, it can be determined that the number of views and the number of comments have a high relationship with sales volume. In addition, if a specific evaluation factor increases but the change in sales volume is insignificant, it can be determined that the relationship between the specific evaluation element and sales volume is low.

In the step of grouping the evaluation factors according to the degree of relationship (S320), evaluation factors having a high degree of relationship may be grouped into one group. For example, as shown in II of FIG. 1, among the evaluation factors (A data, B data, C data, D data), the (A, B, C) group having a high degree of relationship, (B, C, D) Group and (A, C, D) groups can be created and analysis can be performed for each group.

In step S330 of arranging the evaluation elements within the group according to the weight, base data is selected from among the evaluation element data within the group, weights are assigned to each data, and the data is arranged according to the ranking.

The base data selection criterion can be made in various ways.

According to an embodiment of the present invention, an evaluation factor having the largest change in the measurement interval is selected as base data, and the change width is matched by multiplying other evaluation factors by a conversion factor. This will be described in more detail with reference to FIG. 9C below.

In the step of graphing and displaying the rate of change of evaluation factors (S340), the data collected and analyzed in the previous steps are displayed as a graph.

For example, the system may graph the numerical values of evaluation factors as shown in the graphs of FIGS. 9A to 9I . In addition, the system may generate a polynomial trend line graph of the numerical graphs of these evaluation factors, and after creating a differential graph of the polynomial trend line graphs, check the inflection point value of the polynomial trend line graph. Using the slope values or inflection point values of the graphs checked in this way, it is possible to check the rate of change of the evaluation factors, the trend of change, and the correlation between changes.

In the step of generating a demand forecasting graph by merging the graph of evaluation factors with high weight with the graph of evaluation factors with low weight (S350), the scale of the graphs is adjusted by calculating conversion factors after setting comparison standards for evaluation factors. Align and merge graphs using polynomial trend line graphs, differential graphs, and inflection point data, and create a demand forecast graph.

To this end, the step of generating a demand forecast graph by merging the graph of evaluation factors with a high weight with the graph of evaluation factors with a low weight (S350) includes the step of expressing the data of evaluation factors as a time series graph (S351), and the time series graph Converting to a polynomial trend line (S352), creating a differential graph of the polynomial trend line (S353), synthesizing the polynomial trend line of evaluation factor A and the trend line of evaluation element B after the inflection point occurs (S355) , creating a demand forecasting graph based on the synthesized trend line (S356), and the like.

In the step of predicting demand for a product (S400), future demand for a specific product is predicted by using the demand forecast graph generated in step S300.

The step of predicting demand for a product (S400) includes outputting demand forecast data of the selected product through a demand forecast graph (S410), comparing demand forecast data of the selected product and similar products (S420), and It may include a step of correcting the demand forecast data of the product (S430), and the like.

The steps described above do not necessarily have to be performed in the order in which they are written. In addition, it is possible to omit a specific step or to further include other steps as needed.

8 is a diagram showing examples of fields in which BAFP according to the present invention can be utilized. Referring to FIG. 8 , variable data classification methods and correlation weight adjustment methods for collecting necessary information from third-party homepage sites may be implemented in various ways.

For example, the figures extracted through 1) time, day, and season distribution analysis, 2) rise and fall change analysis, 3) sporadic sudden change analysis, 4) sales distribution analysis, 5) market share analysis, etc. Based on this, it is possible to provide an automatic reporting evaluation report for overall items through different visual analysis.

Hereinafter, a method of predicting product demand in this system will be described in detail through specific examples. In the following embodiment, in order to predict future sales volume or future demand for products, 'likes', 'comments', 'average rating', and 'views' data among relevant evaluation factors are used to predict demand graphs. Let's see how to write

Referring to Table 1 below, changes in the number of likes, number of comments, average rating, and number of views related to a specific product can be confirmed. In Table 1, the numerical values were recorded by dividing the interval into 12 sections. In the embodiment of the present invention, one section is defined in units of one month, but it is not necessary to limit one section in units of one month, and if necessary, the section is defined in units of one day, one week, one month, etc. It is also possible to extend or shorten the period.

In one embodiment of the present invention, it is preferable to use only evaluation factors belonging to one task (see FIG. 2), but it is also possible to use the total sum of the same evaluation factors belonging to different tasks or to use the average value thereof. .

For example, it is desirable to analyze future demand using the number of likes, comments, average ratings, and number of views collected through the Naver Shopping site. It is also possible to collect and analyze the number, rating average, number of hits data, or data corresponding to these data.

9A is a diagram illustrating a step of generating an original graph in order to generate a demand forecasting graph. 9a represents the numerical values listed in Table 1 as a line graph.

Referring to FIG. 9A, the number of likes is recorded as the lowest value of 15 times in the first survey section (section 1) and the highest value of 31 times in the 6th section. The difference between the lowest and highest number of likes is 16 times.

The number of comments is recorded as the lowest value at 8 in section 1, and the maximum value at 16 in section 4. The difference between the lowest and highest values for the number of comments is 8.

The rating average is recorded as the lowest value with 4.5 points in the 1st section, and the lowest value with 4.8 points in the 4th section. The difference between the lowest and highest values of the rating average is 0.3 points.

The number of views is recorded as the lowest value with 66 times in the 2nd section, and the highest value with 81 times in the 7th section. The difference between the lowest and highest number of views is 15 times.

As such, since the change values of the evaluation factors are different, it is difficult to directly compare them. Therefore, in the present invention, the rate of change of the evaluation factors is checked after matching the ranges of change by multiplying the evaluation factors by the conversion factor.

In this embodiment, an evaluation factor having the largest change range is selected as base data. The weight value (conversion factor) is multiplied by other evaluation factors to match the change range of the evaluation factor corresponding to the base data. Table 2 below describes conversion factor values for each evaluation factor.

evaluation factor width of change conversion factor number of likes 16 One number of comments 9 1.778 average rating 0.3 53.333 views 15 1.0667

9B is a diagram illustrating a step of generating an initial value matching graph according to a magnification conversion to generate a demand forecasting graph, and FIG. 9C describes a step of generating a rise and initial value correction graph to generate a demand forecasting graph. It is a drawing to

As shown in FIGS. 9B and 9C , the system of the present invention converts the magnifications of the collected data and compares them, and then compares them by matching the rising widths.

Referring to FIG. 9B, the system matches the initial values by multiplying each evaluation factor by a specific value. For example, since the initial value of comments is 8, multiply by 1.875 to convert to 15. Since the initial rating average is 4.5, multiply by 3.333 to convert to 15. Since the initial value of views is 67, multiply by 0.224 to convert to 15. .

By converting in this way, it is possible to check the change trend of each evaluation factor after a specific point in time (initial value). However, in this case, it is difficult to synthesize graphs of evaluation factors, only the trend of evaluation factors can be grasped. Therefore, it is necessary to match the range of change of the evaluation factors.

Referring to FIG. 9C, the system generates data shown in Table 3 below by multiplying each evaluation factor by the conversion factor calculated in Table 2.

After that, the system performs correction to match the initial values so that the change trend of the evaluation factors can be confirmed after the initial values. At this time, since the range of change must be kept the same, the initial value is matched through sum or difference in this step.

Table 4 below shows data set to an initial value of 15 by adding or subtracting a specific value to the number of comments, average rating, and number of views in Table 3 above.

If a graph is drawn using the data in Table 4, a graph as shown in FIG. 9c can be obtained.

9D is a diagram illustrating a step of generating a polynomial trend line graph in order to generate a demand forecast graph.

When the trend line graph of each evaluation factor graph of FIG. 9c is drawn, it is shown in FIG. 9d. In FIG. 9D, for convenience, a 6th order polynomial trend line graph of each evaluation factor is checked.

According to this, the formula for the 6th order polynomial trend line of the number of likes is as shown in Equation 1 below, the formula for the 6th order polynomial trend line for the number of comments is as shown in Equation 2 below, and the formula for the 6th order polynomial trend line for the rating average is as shown in Equation 3 below. , The formula for the 6th order polynomial trend line of hits is shown in Equation 4 below.

Looking at FIG. 9D and the above equations, each graph is similar to each other, but increases and decreases are repeated at different points. In other words, the evaluation factors show a similar trend as a whole, but when examined in detail, some factors show a tendency to increase while other factors show a tendency to decrease at a specific point in time.

In the present invention, future demand is predicted by reflecting the influence of these mutually related evaluation factors on other evaluation factors.

Specifically, after setting the evaluation factors that have the greatest impact on demand as base data, the rate of increase of the next priority evaluation factors is reflected. In this way, when a trend line is drawn by sequentially reflecting the rate of increase of all evaluation factors, the overall trend can be accurately predicted rather than predicted using only one data.

To this end, it is necessary to check the point at which the rate of change changes from positive to negative and the point at which the rate of change changes from negative to positive in each graph. The point at which the rate of change changes from positive to negative (or when the rate of change changes from negative to positive) is a critical point in time when consumers' interests change rapidly, so it needs to be reflected in demand forecasting.

9E is a diagram illustrating a step of generating a differential graph of a polynomial trend line in order to generate a demand forecast graph. The system analyzes the polynomial differential graph to identify the inflection points of the trend line.

The formula for the differential trend line for the number of likes is as shown in Equation 5 below, the formula for the differential trend line for the number of comments is as shown in Equation 6 below, the formula for the differential trend line for average ratings is as shown in Equation 7 below, and the formula for the differential trend line for the number of views is as shown in Equation 6 below. Same as Equation 8.

Referring to Equations 5 to 8 and FIG. 9E, the inflection points of each graph are shown in Table 5 below.

evaluation factor inflection point number of likes 5.678 (decreasing) number of comments 4.802 (decrease), 9.133 (increase) average rating 10.651 (increase) views 6.512 (decrease)

9F is a diagram illustrating a step of generating a primary merged graph in order to generate a demand forecasting graph. According to an embodiment of the present invention, the system generates a primary merged graph using the inflection point data identified above.

Referring to FIG. 9B , the system sets the weight order in the order of number of likes, number of comments, number of views, number of views, and average ratings of number 1 and number 2, respectively.

In the primary merged graph, the graph is created by merging the 4th rank data with the 3rd rank data.

As can be seen in Table 5 above, the 4th rank average data changes significantly after 10.651. Specifically, it can be seen that the rate of change of the average score changes from negative (decreasing trend) to positive (increasing trend) at 10.651. Since this change in rating average trend can pre-reflect changes in the number of third-ranked views or affect changes in the number of views, the system reflects the rating average data in the graph of views in the section after 10.651.

Since the purpose of this embodiment is to generate the final trend prediction graph, it is important to understand the change trend of the graph. To this end, in this embodiment, the trend line is corrected by calculating the average value of the third and fourth rank data in the interval after 10.651.

A primary merged graph reflecting this is shown in FIG. 9F.

Referring to FIG. 9F , it can be seen that the decrease rate of the graph after the first merge is lower than that of the graph of the number of hits before the first merge. In other words, it can be confirmed that the predicted value for future demand increased when the rating average data was reflected than when predicting future demand using only the number of hits data.

Even in qualitative analysis, if buyers' evaluation of a particular product improves, the number of consumers who select the product after looking at rating reviews will increase, which can lead to an increase in the number of likes. It can be considered more accurate than data.

9G is a diagram explaining a step of generating a secondary merged graph in order to generate a demand forecasting graph. According to an embodiment of the present invention, the system generates a secondary merged graph using the primary merged data generated above.

The second-order merge graph is created by merging the first-order merge data with the second-order number of comments data.

As can be seen in Table 5 above, the third-ranked hit data changes significantly after 6.510. Specifically, it can be seen that the rate of change in the number of views changes from plus (increasing trend) to negative (decreasing trend) at 6.510. Since this change in the number of views can pre-reflect the change in the number of second-ranked comments or affect the change in the number of comments, the system reflects the number of views data in the graph of the number of comments in the section after 6.510. For more accurate prediction, it is preferable that the hit count data reflected at this time is data corrected by the 4th rank data (the primary merged data generated above).

A secondary merged graph reflecting this is shown in FIG. 9G.

Referring to FIG. 9G , it can be seen that the range of change in the graph of the number of comments after the secondary merge is reduced compared to the graph of the number of comments before the secondary merge. In other words, when forecasting future demand using only comment count data, demand is expected to increase rapidly, but when forecasting with view count and rating average data reflected, future demand is expected to increase moderately.

Even in qualitative analysis, if the number of comments has recently increased, but the number of views has turned to a decreasing trend, it is not desirable to expect a rapid increase in demand in the future. In other words, from a qualitative point of view, it can be confirmed that forecasting future demand through secondary merged data will be more accurate than predicting demand based on the number of comments alone.

9H is a diagram explaining a step of generating a tertiary merged graph in order to generate a demand forecasting graph. According to an embodiment of the present invention, the system generates a tertiary merged graph using the 2nd order merged data generated above.

The third-order merge graph is created by merging the second-order merge data with the first-order number of likes data.

As can be seen in Table 5 above, the data on the number of second-ranked comments has a large change in the values of 4.802 and 9.133. Specifically, it can be seen that the rate of change in the number of comments changes from positive (increasing trend) to negative (decreasing trend) at 4.802, and changes from negative (decreasing trend) to positive (increasing trend) at 9.133.

Since this trend change in the number of comments can reflect the change in the number of first-ranked likes in advance or affect the change in the number of likes, the system reflects the number of comments data in the graph of the number of likes in the intervals after 4.802 and 9.133. For more accurate prediction, it is desirable that the data on the number of comments reflected at this time be data corrected by the 3rd and 4th ranking data (secondary merged data generated previously).

A cubic merged graph reflecting this is shown in FIG. 9H.

Referring to FIG. 9H , it can be seen that the graph of the number of likes after the 3rd merge has a decreasing trend compared to the graph of the number of likes before the 3rd merge. In other words, when predicting future demand with only the number of likes data, demand is expected to decrease, but when forecasting with the number of comments, number of views, and average rating data reflected, it can be predicted that future demand will change to an increasing trend.

Even in qualitative analysis, if the number of likes is still declining but all other evaluation factors (number of comments, average rating, number of views) are shifting to an increasing trend, it is desirable to expect that the number of likes will also shift to an increasing trend in the future. In other words, from a qualitative point of view, it can be confirmed that forecasting future demand through tertiary merged data will be more accurate than predicting demand based on the number of likes alone.

9H is a diagram illustrating a method of predicting future demand through a trend prediction graph.

The system generates a trend prediction graph based on the cubic merged graph created in the manner described above.

In the trend prediction graph, a trend prediction graph is generated using data from the section immediately before the inflection point (section 4) to the latest section (section 12), and values 2 sections after the latest section are predicted.

The trend prediction graph is created as a polynomial trend line graph, and the degree of the polynomial is '(the number of inflection points of the second-order data) + 1'. Since the number of inflection points refers to the number of x values at which the differential value of the graph becomes 0, the degree of the graph is preferably '(the number of inflection points of the second-order data) + 1', which is the value obtained by integrating it.

Therefore, in this embodiment, the trend prediction graph is composed of a 3rd order polynomial trendline graph, and according to this 3rd order polynomial trendline graph, the y value is approximately 26 in 14 intervals.

The system presents a reporting result to the user using the demand prediction result.

10 is a diagram showing an example of reporting based on demand prediction results thereafter.

In FIG. 10, the up/down values are calculated as (predicted section value/recent section value).

In this embodiment, the value of the predicted section (section 14) is 26, and the value of the latest section (section 12) is 20.3, so the value of the up/down is about 1.28.

Referring to FIG. 10, the system provides a report to the user, 'This item is appropriate for the trend but requires a careful approach due to its weak growth potential' corresponding to the up/down figure of 1.28.

The analysis example reporting statement described in FIG. 10 is only an example for explaining the present invention, and it is possible to change specific phrases as needed.

According to at least one embodiment of the present invention described above, it is possible to increase the utilization as a market analysis model by predicting market demand by analyzing data acquired in real time from the market in operation in real time, and to utilize a large amount of data. The designed big data analysis module (BAFP) enables accurate analysis and increases the target scalability of big data analysis, enabling analysis for various industries and countries, and analysis by generation and country according to consumption patterns. It enables customized demand forecasting according to each target using intergenerational behavior type analysis and overseas consumption patterns, ranks multiple factors influencing demand, and changes the rate of change of low-ranking factors to the rate of change of high-ranking factors By merging with, it is possible to more accurately predict the rate of change of demand, etc., and quantitatively/qualitatively improved effects can be expected compared to the prior art.

The embodiments described in this specification and the accompanying drawings merely illustrate some of the technical ideas included in the present invention by way of example. Therefore, since the embodiments disclosed herein are intended to explain rather than limit the technical spirit of the present invention, it is obvious that the scope of the technical spirit of the present invention is not limited by these embodiments. All modifications and specific examples that can be easily inferred by those skilled in the art within the scope of the technical idea included in the specification and drawings of the present invention should be construed as being included in the scope of the present invention.

100: big data analysis module
110: data collection unit
111: communication module
112: database
120: data analysis unit
130: demand forecasting unit
140: management department

Claims (10)

a data collection unit that selects evaluation factors that affect product demand and collects data of evaluation factors;
a data analysis unit that recombines the evaluation factors by calculating the relationship and weight of the evaluation factors; and
A demand forecasting unit for predicting future demand for products based on the data recombined by the data analysis unit;
The data analysis unit calculates the relationship between the selected evaluation factors, groups the evaluation factors based on the calculated relationship, sets the ranks of the grouped evaluation factors, and evaluates the evaluation elements with high rankings to those with low rankings. Merge element data to create a demand forecast graph;
The ranking of the evaluation factors is,
Within the set interval range, the higher the difference between the maximum and minimum values of the data, the higher the rank.
The data analysis unit,
After multiplying the data of other evaluation factors except for the highest ranking (ranking 1) evaluation factor by the conversion factor to match the range of change of the data with the highest ranking (ranking 1) data, the evaluation elements in the first section have the same value. Consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that for calibrating. delete According to claim 1,
The demand forecasting unit,
Big data analysis framework characterized in that formed to correct the demand forecast data of the selected product by outputting demand forecast data of the selected product through the demand forecast graph and comparing demand forecast data of the selected product and similar products. A consumption pattern analysis and future demand forecasting system using the platform. delete delete According to claim 1,
The data analysis unit,
After estimating the polynomial trend line of the evaluation factors, creating a differential graph of the polynomial trend line to determine the location of the inflection point, and merging the data of the evaluation factors based on the location of the inflection point to create a demand forecast graph. Characterized in that Consumption pattern analysis and future demand forecasting system using big data analysis framework platform. According to claim 6,
The demand forecasting unit,
Consumption pattern analysis using the big data analysis framework platform, characterized in that future demand is predicted based on the ratio of the value of the first evaluation factor predicted by the demand prediction graph and the value of the recently measured first evaluation factor and future demand forecasting systems. According to claim 6,
Data merging of the evaluation factors,
The inflection point of the trend line of the second-ranked (second-ranked) evaluation element is identified, and in the intervals after the inflection point, the data of the first-rank evaluation element is replaced with the average value of the data of the second-rank evaluation element and the data of the first-rank evaluation element. Consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that for doing. According to claim 8,
The demand forecast graph,
Consumption pattern analysis and future demand forecasting using the big data analysis framework platform, characterized in that it is drawn as a polynomial trend line from the section immediately before the inflection point to the section after the recently measured section among the data of the merged first-rank evaluation factors. system. According to claim 9,
The demand forecast graph,
Consumption pattern analysis and future demand forecasting system using a big data analysis framework platform, characterized in that the polynomial trend line of the order of adding 1 to the number of inflection points of the second-order evaluation factor.

Images