KR20030003397A - Method of content catrgory for data analysis - Google Patents

Abstract

1. TECHNICAL FIELD OF THE INVENTION

The present invention relates to a content classification method for data analysis and a computer readable recording medium having recorded thereon a program for realizing the method.

2. The technical problem to be solved by the invention

The present invention provides a content classification method for designating and storing a content classification code that can be efficiently used for large-scale data analysis such as data warehousing, data mining, user disposition analysis, and customer relationship management (CRM). To provide a computer-readable recording medium for recording the program.

3. Summary of Solution to Invention

According to an aspect of the present invention, there is provided a content classification method for data analysis applied to a content classification system, wherein the content is classified into a horizontal type according to a content classification characteristic, and the content is classified into a vertical type until each classified content class is completely classified. Stage 1; A second step of calculating the total number of bits required for the content classification by adding the required number of bits for each horizontal level of the level graph and the number of bits required in each horizontal node constituting the level; And a third step of allocating the bits from the highest horizontal level by designating a code of each classification item for the corresponding bits when the total number of bits required is calculated.

4. Important uses of the invention

The present invention is used for a content classification system.

Description

Methods of content categorization for data analysis

The present invention relates to a content classification method for data analysis in a content classification system and a computer-readable recording medium recording a program for realizing the method, in particular, online shopping, customer relationship management system, data mining, data ware. Content classification code to code and classify content and store it with minimal storage capacity in systems that need to store, search, operate, and analyze large-scale content such as housings, Internet broadcasts, digital libraries, and video on demand systems And a computer readable recording medium having recorded thereon a program for realizing the method.

In most cases, a method of classifying content is to classify the content into large categories, and then classify the items in detail and specify the classification codes in integer or character types. In this method, there is a lot of difficulty in analyzing content usage such as user disposition analysis because it is inconsistent among the classification code designation methods and it is difficult to show the association between various contents. As the number of contents increases, the storage capacity required increases. There was a problem.

The present invention has been proposed to solve the above problems, and designates a content classification code that can be efficiently used for large-scale data analysis such as data warehousing, data mining, user disposition analysis, and customer relationship management (CRM). And a computer-readable recording medium recording a content classification method for storing and a program for realizing the method.

1 is an exemplary configuration diagram of a hardware system to which the present invention is applied.

2 is an exemplary diagram of a content classification code designation method in units of bits to which the present invention is applied;

3 is a diagram illustrating an embodiment of a content classification method for data analysis according to the present invention;

4 is a level graph for calculating the number of bits required in the content classification method according to the present invention.

5 is an exemplary diagram of a bit configuration design for content classification and storage in a content classification method according to the present invention;

6 is a flowchart illustrating a content classification method for data analysis according to the present invention.

* Explanation of symbols for the main parts of the drawings

11: central processing unit 12: main memory unit

13: auxiliary memory device 14: input device

15: output device

The present invention for achieving the above object, in the content classification method for data analysis applied to the content classification system, according to the content classification characteristics, the content is classified horizontally, until each classified content class is completely classified A first step of classifying as vertical; A second step of calculating the total number of bits required for the content classification by adding the required number of bits for each horizontal level of the level graph and the number of bits required in each horizontal node constituting the level; And a third step of allocating the bits from the highest horizontal level by designating a code of each classification item for the corresponding bit when the total number of necessary bits is calculated.

In addition, the present invention provides a content classification system having a processor, comprising: a first function of classifying content in a horizontal form according to content classification characteristics and classifying the content in a vertical form until each classified content class is completely classified; A second function of calculating the total number of bits required for the content classification by adding the required number of bits for each horizontal level of the level graph and the required number of bits in each horizontal node constituting the level; And a computer-readable recording medium having recorded thereon a program for realizing a third function of allocating bits from the highest horizontal level by designating a code of each category item for the corresponding bits when the total number of required bits is calculated.

The present invention can be usefully used to classify contents and classify not only multimedia contents but also user information, product information, and the like of the contents, and as a result, a large amount of data is used to store a large amount of data. So that it can be stored effectively. In addition, the fast data can be analyzed by bit operation, so it can be efficiently used for tasks such as customer relationship management, data warehouse, and data mining that require large-scale data analysis.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is an exemplary configuration diagram of a general hardware system to which the present invention is applied.

As shown in FIG. 1, a general hardware system includes a central processing unit 11, a main memory device 12 connected to the central processing device 11, and an auxiliary memory device 13 connected to the main memory device 12. And an input device 14 and an output device 15 connected to the central processing unit 11.

Here, the hardware system, the central processing unit 11 for controlling and managing the overall operation of the computer, the program stored in the central processing unit 11 and stores a variety of data used during or during operation And a main memory device 12, an auxiliary memory device 13, and input / output devices 14 and 15 for inputting and outputting data to and from a user.

The auxiliary memory device 13 stores a large amount of data, and the input / output devices 14 and 15 include a general keyboard, a display device, and a printer.

In the main memory device 12 of the hardware system as described above, a program for designating and storing a content classification code that can be efficiently used for large-scale data analysis, such as data warehousing, data mining, user disposition analysis, and customer relationship management (CRM). Is stored and performed under the control of the CPU 11.

In the present invention, a content classification method will be described using a simple Voice On Demand (VOD) system as an example in order to classify content into a predetermined number of bit arrays as shown in FIG. 1.

First, content is classified according to content classification characteristics.

(1) The content is classified into a horizontal type. Horizontal classification refers to the characteristics that all the content has at the same time, such as the highest category, the country of origin, the size of the content, and the usage price.

For example, if the content of the VOD system is classified horizontally, it can be classified into "production country" and "big classification".

(2) It is classified vertically for each class classified in (1) above. Vertical classification means reclassifying the same kind of content in detail.

For example, the category "Production Country" is classified into five countries: "Korea", "United States", "Japan", "China", and "France", and the category "Classification" is "Movie" and "Sports". , "Education".

(3) If each item listed in (2) above is not an item that can no longer be categorized, the process is repeated again from (1) until it can no longer be classified for each item.

For example, items such as "Korea" and "United States" that vertically categorize "production country" need not be classified any more. However, the horizontal classification of (1) is again performed for "movie" which is a sub item of "major classification". In other words, if you categorize the characteristics that all "movies" have in common, you can classify them horizontally into "grade", "production form", "classification by content", etc., and "grade" is not "under 18 years old". "," Can not be viewed under 12 years old "and the like. In addition, "production format" is vertically categorized into "general shooting", "special effect shooting", "cartoon movie", "three-dimensional animation", etc., "classification by content" is "romantic comedy", "action", Vertically classified as "fear", "SF", and the like. Therefore, in FIG. 3, the task of classifying the content in this manner is represented by a graph.

Second, calculate the number of bits needed for content classification.

When the content is classified as described above, the number of bits capable of storing the content is calculated. Here, in the content classification method of the first process, a top-down design technique for subdividing contents from a higher level to a lower level is used. However, a bit storage design is used from a lower-level to a higher-level bottom-up design technique.

That is, to calculate the number of bits required for content classification, a level graph as shown in FIG. 4 is drawn. Each level consists of a horizontal classification and a lower level vertical classification.

Horizontal classifications are all rectangular, vertical classifications are circular if they can only have an OR relationship, and hexagons if they can have an AND relationship (for example, a movie's rating is "Audience" or " Must be at least 12 years old "or" Over 18 years old "At the same time, both items are not possible, so in this case only the OR relationship is possible. However, the contents of the" movie "can be" action "and" SF ". In this case, an AND relationship is possible).

Therefore, the total number of bits required for content classification is the sum of the number of bits required for each level, and the number of bits required for each level is the number of bits required for each item (represented by a node in the graph) of the horizontal classification level. . That is, the total number of bits is the same as the sum of all the number of bits required in the horizontal classification item represented by the rectangle in the level graph as shown in FIG.

On the other hand, the number of bits required for each item of each horizontal classification level is determined when the lower vertical classification item is only capable of OR relationships (i.e., when the lower element of the graph is circular). If the lower vertical classification items have an AND relationship (that is, the lower element in the graph is hexagonal), the number of vertical classification items is obtained. For example, three bits are required to represent five items of the country of origin. ).

However, at present, only five countries are needed, but the content of other countries can be included later. Up to 256 countries can be marked by allocating 8 bits to represent all the countries of the world. For example, assuming that the contents of a movie are classified into "romantic comedy", "action", "fear", and "SF", four bits are required. In case of "action" and "horror film", it can be expressed as 0110, and in case of "SF" film, it is represented as 0001.

That is, the minimum number of bits for representing the content classification code is calculated as described above, but considering the scalability, a sufficient number of bits is specified by predicting the number of items that can be added later. For example, although only five countries are considered as production countries, content of a country that is not included in the present country can be included in the system, so 8 bits can be designated instead of 3 bits to ensure future scalability.

Third, do bit structure and code design.

That is, when the required number of bits is calculated, it is determined how to use the corresponding bits and how to specify each classification item as a code. In the example of FIG. 4, all 13 bits are required. To properly configure this, bits are allocated from the highest horizontal level again using the top-down technique. 5 is a graphical representation of such a bit structure.

In the case of the vertical AND item, meaning is given for each bit, and in the case of the vertical OR item, several bits are gathered to have one meaning.

Referring to the operation of the content classification method for data analysis of the present invention having the structure as described above in detail.

6 is a flowchart illustrating a content classification method for data analysis according to the present invention.

As shown in FIG. 6, first, in order to classify the content for data analysis according to the present invention, the content is classified horizontally (601), and vertically classified into the horizontal classes for each class (602).

Subsequently, it is checked whether each class categorized vertically is completely classified (603), otherwise, the process is repeated from the process of classifying the content horizontally (601), and a level graph is drawn (604) if it is completely classified.

Next, for each horizontal node, it is determined whether the lower vertical nodes are in an AND relationship or an OR relationship (605). If the AND relationship is required, the number of bits required is equal to the number of vertical classification items (606). Determine (607).

Then, the required number of bits for each horizontal level of the graph is calculated by adding up the required number of bits in each horizontal node constituting the level (608), and the total number of necessary bits is determined as the sum of all necessary bits in each level. (609).

Finally, in order to store the content classification code, the calculated number of bits is allocated and bits are allocated from the highest horizontal level using the level graph for each bit (610).

The method of the present invention as described above may be implemented as a program and stored in a computer-readable recording medium (CD-ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.).

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and various substitutions, modifications, and changes are possible in the art without departing from the technical spirit of the present invention. It will be apparent to those of ordinary knowledge.

As described above, the present invention can achieve the following effects by bit-coding, classifying, and storing contents.

First, there is an effect that can minimize the disk efficiency. For example, when all classification information is designated and stored as an integer code, 32 X 5 bits are required to store large classification information, country information, and three levels of small classification information. However, when using the technique of the present invention, 32 bit codes are used. It takes only 32 bits.

Secondly, if the required number of bits is given a margin, it is possible to process even when the number of classifications increases, thereby increasing the scalability. For example, if 10 major classifications are assigned, 4 bits can be designated for the major classification code, so 16 major classifications can be specified, and up to six additional major classifications can be specified.

Third, in general, bit operations operate much faster than integer operations or string operations, so when analyzing data using classification codes, it is possible to guarantee much faster performance than other techniques. For example, to process the query "What percentage of the content the user used during the month of May 2001 was Korea," simply set the country of production bit in the content classification code in the user log file. You can do AND with 0010. If you represent this information as a string, you must perform a number of string comparisons.

Finally, there is an effect that can express the association between different content. For example, if the country code is designated as 8 bits from 4th to 11th bits for all contents, the content classification by country becomes possible.

Claims (4)

In the content classification method for data analysis applied to the content classification system, A first step of classifying the content horizontally according to the content classification characteristic and classifying the content vertically until each classified content class is completely classified; A second step of calculating the total number of bits required for the content classification by adding the required number of bits for each horizontal level of the level graph and the number of bits required in each horizontal node constituting the level; And A third step of allocating bits from the highest horizontal level by designating a code of each classification item for the corresponding bits when the total number of required bits is calculated; Content classification method for data analysis comprising a. The method of claim 1, The total bit number calculation process, The number of bits required at each level is added, and the number of bits required at each level is calculated by adding the number of bits required for each item of the horizontal classification level. The method according to claim 1 or 2, The number of bits required for each horizontal level is If the subordinate vertical items are OR relationships And the number of vertical classification items when the lower vertical classification items are in an AND relationship. In a content classification system having a processor, A first function of classifying the content horizontally according to the content classification characteristic and vertically classifying the classified content categories until they are completely classified; A second function of calculating the total number of bits required for the content classification by adding the required number of bits for each horizontal level of the level graph and the required number of bits in each horizontal node constituting the level; And A third function of assigning a bit from the highest horizontal level by designating a code of each classification item for the corresponding bit when the total number of required bits is calculated A computer-readable recording medium having recorded thereon a program for realizing this.