KR20020084328A - a mode of life structure analysis method of internet sites - Google Patents

Abstract

PURPOSE: A method for analyzing a life mode structure of an Internet site is provided to analyze a life mode structure of an Internet site through investigating visitors of the Internet site. CONSTITUTION: A plurality of Internet users(10-1-10-n) connects to a wanted Internet site through an Internet network(1). The Internet users(10-1-10-n) obtain one's wanted information and supply one's possessed information to other person through not only one site but also many Internet sites. The entire structure of the Internet is analyzed based on a population statistical characteristic of users using a result of an investigation based on Internet users by thinking the Internet as one ecosystem. As a method for investigating Internet users, access methods based on a site, an advertisement, and a visitor are provided.

Description

{A mode of life structure analysis method of internet sites}

The present invention relates to a method for analyzing the ecological structure of an internet site, and more particularly, to a method for analyzing the ecological structure of an internet site through an internet site visitor survey.

In recent years, the Internet is rapidly developing as a driving force to accelerate change in all aspects of our lives. The Internet is creating a new economic and management paradigm economically, politically suggesting the possibility of cyber democracy and vitalizing the populace, and socially, showing the prospect of a new community revival. But the Internet is also bringing us many new problems at the same time. Problems such as hackers and information security, surveillance and privacy violations, social norms and netiquette, and intellectual property disputes are unthinkable before the advent of the Internet. Thus, there is a growing need for social science research on the Internet that is rapidly penetrating into our lives and creating amazing change.

However, until now, research on the Internet has been mainly focused on the qualitative aspects and on the personal observations. Therefore, more and more objective and comprehensive empirical studies on the Internet are urgently needed, with more than half of the people using the Internet through the spread of the Internet.

Accordingly, an object of the present invention is to provide a method for analyzing the ecology of an internet site.

Another object of the present invention is to provide a method for identifying the ecological status of an Internet site within an ecological structure.

In order to achieve the above objects, in the present invention, using the site-centered, advertising-centered and visitor-centered approach to investigate the characteristics of the Internet users, the difference indices of user characteristics, overlapping index of user characteristics and user overlapping By examining the relationship between the Internet sites using the index, it provides a method for analyzing the ecological structure of the Internet site characterized by analyzing the ecological status of the Internet site.

1 is a schematic configuration diagram showing an internet connection state.

2A and 2B are site distribution diagrams according to difference indices for explaining the present invention.

3A and 3B are distribution charts according to user property overlapping index for explaining the present invention.

4A and 4B are distribution diagrams according to a user overlap index for explaining the present invention.

5 is a scatter diagram illustrating the relationship between the average user overlap index and visitor increase / decrease for each site for explaining the present invention.

6A is a regression line showing overlapping indexes with other type sites for explaining the present invention, and FIG. 6B is a regression line showing overlapping indexes with the same type sites.

[Simple explanation of symbols for main parts]

1: Internet network 10-1 to 10-n: Internet user

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic configuration diagram showing an internet connection state.

Referring to the drawings, a number of Internet users 10-1 to 10-n connect to their desired sites through the Internet network 1. The Internet users may obtain information they want through not only one site but also a plurality of Internet sites, and provide information that they own to others. On the Internet (1), Daum, portal sites such as DreamWiz, DreamX, etc., search engine sites such as Yahoo, Naver, Lycos, Altavist, etc., and Internet service provider (ISP) sites such as Hananet, Durnet, etc., SBS, Broadcast sites such as KBS, MBC, newspaper sites such as Chosun Ilbo, Hankyoreh, JoongAng Ilbo, Dong-A Ilbo, Hankook Ilbo, mobile phone sites such as n016, speed011, 018, chat site such as sayclub, skylove, dearyou, send2u There are various sites, such as card sending sites, etc., which can be accessed anytime, anywhere 24 hours.

The present invention is characterized by analyzing the overall structure of the Internet based on the demographic characteristics of the user with the result of the Internet user-centered survey, which is a kind of empirical research on the Internet. The theoretical starting point of the present invention is to view the Internet as an ecosystem. The sites and users that make up the Internet are combined by the agreement of their interests and tastes, but the demographic characteristics of users exist in the background of the interests and tastes. In other words, it constitutes an ecological environment that must be adapted in the long run.

The relationship between Internet sites, which are mediated by the demographic characteristics of users, is an important basis for the strategic planning of Internet sites. The distribution of demographic characteristics of users constitutes a social context in which Internet sites are embedded, and the possibility of competition and cooperation of Internet sites is conditional on this social context. In this sense, there is no reason for the interaction to be random in the Internet, which is generally considered as a space of anonymity, and there is a sufficient possibility of users' peculiar species even when the possibility of direct interaction is limited. Starting from these assumptions about the social embryonicity of the Internet, we use empirical data to confirm the distribution of the niche of the internet site, and how the degree of overlapping ecological status is achieved through competitive and symbiotic interactions. Analyzes whether it affects performance.

First of all, the survey method for Internet users is largely mobilized by site-based approach, advertisement-oriented approach, and visitor-oriented approach.

First, site-centric measurements are based on log files of Internet sites. This approach is the most common method used by Internet sites, and log files that track the IP addresses of these users through cookie files left on the visitor's computer when accessing the Internet site. Measure if requested. Since this site-based measurement method is a complete survey, there is little room for sampling error, but there is room for measurement error. This is because different sites have different ways of linking pages, and there are technical elements that make accurate measurements difficult, such as frames, subwindows, and Java applets. In addition, site-centric measurement has a problem that it is not easy to compare with other sites. A systematic comparison to other standard methodologies between sites is therefore impossible to obtain in a site-centric approach. At present, KT Internet Corporation is conducting user research mainly on 10 sites.

Secondly, similar to the site-centric approach, the advertisement-centric approach measures people who come to the site, but the standard of measurement is not the site itself, but the third-party ad server. Ed Server is a server site that specializes in providing only advertisements, and is responsible for sending advertisements to people who have visited these sites by linking to sites that advertisers choose to place ads. Advertising-centric user measurement is like visiting the site and analyzing its characteristics. Although this measurement method may be objective compared to measuring log files for each site, there is a technical complexity problem similar to site-centric measurement. In addition, both site-centric measurement and advertisement-centered measurement require a lot of costs for user measurement on many sites, and there is a problem in that cooperation of the sites is required.

Third, the visitor-centered approach is a measure of borrowing ideas from television's rating method. By installing a software meter on the computer of a large number of randomly selected Internet user panels, they record their browsing history, and measure the software's records. In most cases, the meter records the address of the visiting Internet site left on the user's computer. Site- or ad-centric measurements are based on the IP address of a machine, not a person, or a computer used by Internet visitors, making it difficult to determine the specific characteristics of an individual. Analyzing a person's visits to a site can provide a wealth of information and be objective. However, there is room for sampling error because this method uses a sample. In particular, in the case of the Internet, each site becomes a measurement item.

After all, if you're going to research a myriad of sites, you'll need to have a very large sample size, or a huge number of panels, to make a good survey. In particular, increasing the number of panels while maintaining the probability of sampling is expensive. According to a study that compares the results of site-centered measurement with the total survey and the reliability of the measurement, the measurement of the sites ranked at the top of the ranking using the general sample size of the current offline survey is quite accurate but the number of visitors goes to the bottom. Is estimated to be lower than actual.

In this sense, user-centered measurement is essential to the selection and management of panels. In order to correctly select an Internet user panel, it is important not to give a choice. At present, the internationally user-centered measurement agencies are represented by US Media Metrix and Nielson NetRatings, and several domestic companies such as Internet Matrix Korea Metrix are conducting user-centered research.

On the other hand, the measurement of the ecological status of Internet sites using the demographic characteristics of visitors by site was generally performed by the following procedure. First, we analyzed the demographic characteristics of visitors, such as gender, age, education, and occupation, by site. The similarity or difference between the site pairs was compared with the distribution of demographic characteristics by site. We also compared how many users visited each site pair, that is, a person who visited one site visited another site as well. As a result, if you compare the demographic characteristics of the user to n sites, the actual comparison result will be for n (n-1) pairs. The relationship between the sites was estimated with the similarity and heterogeneity in the distribution measured by site pairs, or the degree of duplication of actual visits. There are three indexes used to compare sites.

First, the dissimilarity index in the user characteristics shown below indicates the degree to which the composition ratio of the demographic characteristics is different.

D = 0.5 X ∑∑ | Pik / P-Pjk / P |

In this equation, i and j indicate the two objects to be compared, and Pk / P is the composition ratio of the kth category of the characteristic being compared. This index was developed in human ecology to measure how integrated or segmented the racial or occupational gender composition of the community is. This index indicates the proportion of people who must move in order for the characteristic composition of the populations of the two subjects being compared to be equal. The values of the exponents range from 0 to 1 (or 100), 0 means that the characteristics are completely consistent without the need for movement at all, and 1 (or 100) means that the specification is completely different so that everyone has to move. . This index was used to compare the composition ratio of age, education, gender, and occupation among demographic characteristics of each site pair, and then the four values were summed and used as the overall index of site difference.

Second, the niche overlap index indicates the degree of overlap in the range of ecological status identified by demographic characteristics. This index was developed by Miller McPherson to compare the ecological status of voluntary associations. In accordance with McPherson's suggestion, we identified the age and educational background of the site's visitors. The age range of Site A visitors is measured as an average (1.5 standard deviations), and Site B's age range is measured equally, and the overlapping ranges of these two sites are considered overlaps in the ecological status range of the sites. In the same way, the extent of range overlap in school ages was identified, and the degree of overlap between the age and school age ranges was multiplied and used as a general index of overlap among sites. Here, sites that cover a wide range of age and school age users will have high indexes for many sites and ecological status. These sites can be said to take a comprehensive strategy. On the other hand, sites that focus on users of a certain age and school age will have fewer sites and higher ecological status index. These sites take an expert strategy.

Finally, the index of user overlap was calculated. This index indicates the degree of overlap of users by site. In other words, we measured how many times the same user visited site A and site B in order to more directly grasp the degree of overlap between the sites. These measures are similar to constructing similar networks through data on joint participation in network analysis. After calculating the overlapping index of users between the sites, the final index was standardized by the number of users of the two sites. This standardized index, like other indices, has a translation of 0 to 1, where 0 means that the two sites being compared have no overlap, i.e. completely different users visit, and 1 is completely redundant, i.e. Visiting the site means that they are the only users of these sites.

2A and 2B are site distributions according to the disparity indexes of June and August, respectively, and FIGS. 3A and 3B are distributions according to user characteristic overlapping indexes of June and August, respectively, and FIGS. 4A and 4B are respectively. The distributions are based on the user overlap index in June and August, respectively.

Referring to the drawings, it can be seen that the indexes have similarities with each other despite being different. In other words, the clusters of sites by type are more pronounced in June and somewhat weaker in August. As a result of classifying the sites by comparing only the similarities, differences, or duplications of Internet users, similar types of sites are clustered closely together on the graph. As mentioned above, it can be said that the appearance of the Yu-Yeong-jong species appears in the visit of the site due to the different interests and tastes according to demographic characteristics of users on the Internet. In addition, it can be seen that the characteristics of the group of users that the target sites of yoyojong mainly aim to define the ecological status of these sites.

Then, referring to the drawing, next, portal sites such as DreamWiz and DreamX gather in the center to form a cluster. In addition, search engine sites such as Yahoo, Naver, Lycos, Altavist, etc., and Internet service provider (ISP) sites such as Hananet and Thrunet, and broadcasting sites such as SBS, KBS, and MBC are located in the center. . In the vicinity, newspaper sites such as Chosun Ilbo, Hankyoreh, JoongAng Ilbo, Dong-A Ilbo and Hankook Ilbo are located. Mobile phone sites such as n016, speed011 and 018 are concentrated on one side. In addition, it can be seen that clustering phenomena such as chat clubs such as sayclub and skylove and card sending sites such as dearyou and send2u are clear.

Interesting facts that can be found through the above drawings, a lot of visitors gathered in the I Love School, as the rankings rise rapidly, the user characteristics are also diversified, it can be seen that the younger people moved to the multi-group as the use of a somewhat older age. The dynamics of these alumni community sites may reflect the differentiation according to the aforementioned species and demographic characteristics.

5 is a scatter diagram illustrating the relationship between the average user overlap index and visitor increase / decrease for each site for explaining the present invention.

Referring to the figure, the average degree of user duplication of the site and all other sites without a distinction between competition and symbiosis is averaged, and shows the relationship between the average degree of duplication and the increase and decrease of the number of visitors to the site. If you look at the X-axis, you can see that portal sites have a high degree of redundancy on average. All sites with an average overlap of 0.25 or more are portal sites. On the other hand, the Y-axis shows the largest variation in the number of visitors to business and portal sites. As a result, competition between them is more intense. The relationship between the overlapping index and the increase and decrease of visitors generally shows that the higher the average duplication, the greater the number of visitors, indicating that the overlap of users has a positive effect on the performance of the site. As a result, it can be said that the result of user duplication has increased the number of users through sharing of users rather than being deprived of other users.

6A is a regression line showing overlapping indexes with other type sites for explaining the present invention, and FIG. 6B is a regression line showing overlapping indexes with the same type sites.

First, referring to FIG. 5A, an average overlapping index with other types of sites is obtained to see a relationship with performance. This means the effect of overlap in symbiotic relationships. As in FIG. 4, it can be seen that users visit more sites with high degree of overlap in a symbiotic relationship. Eventually, our expectation that the effect of redundancy through symbiosis will have a positive effect on the performance of sites has gained empirical support.

On the other hand, Figure 5b shows how the average degree of overlap with other sites of the same type resulted in performance. Since sites belonging to the same type are in a competitive relationship, the average degree of duplication and site performance are positive in this case, contrary to our expectation that the higher the degree of duplication, the higher the pressure of competition and the lower the performance. . However, the slope of the regression line is slower than in the case of symbiotic relationship. In addition, except for portal sites, other types of sites generally show that the higher the overlap in competition, the lower the increase.

As such, the visitor characteristics of Internet sites provide important information that determines the ecological status of the sites. This is because Internet users have different interests and tastes according to demographic characteristics, and their interactions and visit patterns on the Internet are strongly influenced by such distinctive interests and tastes. As a result, these peculiar tendencies of Internet users put Internet sites in a structured space with social characteristics and relationships rather than indiscriminate and atypical space. In short, we can see that online life and institutions are influenced by offline and reflect the characteristics of offline rather than being completely independent from offline, which we call social sociality and composition of Internet sites.

In the present invention, how the Internet sites are arranged by the social characteristics of users is presented through a multi-dimensional scale graph. Through these graphs, we can see the clustering and dispersion of Internet sites. It is also a glimpse of sites trying to relocate their ecological status through strategic efforts.

In addition, it can be seen that the characteristics of internet users affect the performance of ecological crisis sites. Internet sites that share the same ecological status and thus have a high degree of overlap have two possibilities. One is the potential for better performance through synergy and cooperation between sites that share ecological status. We see this relationship as a symbiosis in the Internet ecosystem. The other is that competition among sites with overlapping ecological geographies may result in competition for exclusion, resulting in poor performance. This relationship is a competition in the Internet ecosystem. Of course, whether symbiosis prevails or competition prevails is not an arbitrary choice of Internet sites. We can see that the higher the user's duplication, the lower the performance will be due to competition among sites of the same type, based on the type classification according to the function and contents of the Internet site, and overlapping by the logic of symbiosis between different types of sites. The higher the degree, the more likely the relationship is to be achieved.

As described above, the present invention analyzes the ecological structure of the Internet sites by using the visitor characteristics of the Internet sites, and analyzes the ecological status of each Internet site within the ecological structure, thereby making the Internet useful for our lives more active. You can get the effect that you can use.

Claims (5)

Investigate the characteristics of Internet users by using site-based, advertisement-based, and visitor-centered approaches, and examine the relationships among Internet sites by using the differences of user characteristics, the overlapping index of users, and the overlapping index of users. Thereby analyzing the ecological status of the internet site. The method of claim 1, wherein the site-based approach is based on a log file of the Internet site. The method of claim 1, wherein the advertisement-centric approach uses a dedicated advertisement medium. The method of claim 1, wherein the visitor-centered approach is characterized by installing a software meter on the computer of a plurality of randomly extracted internet user panels, which records their internet usage, and using the records left by the software. How to analyze the ecology of an internet site. According to claim 1, The index of dissimilarity of the user characteristics indicates the degree to which the composition ratio for each demographic category is different, the overlapping index of the user characteristics indicates the degree of overlap of the range of ecological status identified by the demographic characteristics, The overlapping index is a method of analyzing the ecological structure of an Internet site, characterized by the overlapping degree of users by site.