KR20060103034A - Internet advertising service system and method thereof - Google Patents

Abstract

The present invention relates to an Internet advertising service system that provides a service for advertisers having a website to advertise their website on the web more efficiently. The system of the present invention collects and analyzes keywords described in web documents and search keywords of an Internet user delivered through a search engine web server, and associates between search keywords of a search engine and keywords described in each of the web documents. Create an association between and build it into a database. The keywords built in the database are recommended to the advertiser and processed so that the advertisement can be displayed on a list of search engines or web documents according to the bidding history of the advertiser for the recommended keywords. The system also performs cost per visitor (CPV) charging. Such a system enables a more advanced keyword recommendation method to purchase more suitable keywords for an advertiser's website, and more efficiently execute an advertising budget.

Ads, keywords, web documents, extractions, referrals, bidding, billing, CPV,

Description

Internet advertising service system and method

1 is a network configuration to which the Internet advertising service system according to the present invention,

2 is a block diagram of an Internet advertising service system according to an embodiment of the present invention;

3 and 4 are diagrams for explaining a keyword collection and suggestion process through a search engine according to an embodiment of the present invention;

5 and 6 are diagrams for explaining a keyword collection and promotion process through web documents according to an embodiment of the present invention;

7 and 8 are diagrams for explaining a keyword bidding process according to an embodiment of the present invention;

9 is a view showing a bidding tool according to an embodiment of the present invention,

10 and 11 are views for explaining an advertisement exposure process on a web document according to an embodiment of the present invention, and

12 is a view for explaining a charging process according to an embodiment of the present invention.

* Description of the symbols for the main parts of the drawings *

110, 120, 130: Internet terminal 150: Internet

200: Internet advertising service system 210: search engine web server

220: web analytics server 230: database server

240: ad server 242: keyword recommendation processing unit

244: bid management unit 246: billing unit

248: advertisement processing unit 510: media next server

520: cafe server 530: other web document server

The present invention relates to an internet advertising service system and method, and more particularly, to an internet advertising service system providing an internet service for advertisers having a website to advertise their websites more efficiently. .

With the development of the Internet, various websites are appearing exponentially around the world. In other words, various websites are being opened for other reasons such as public relations, community means of hobbyists, and individuals voluntary network participation in companies or organizations. As the website appears to be numerous, general Internet users (here, general Internet users are people who use the Internet for simple reasons such as information retrieval, and hereinafter referred to as Internet users) can visit the desired websites. Search sites that serve a variety of appearances. These search sites display countless Internet sites by area, or find and list keyword related sites for keyword input related to the desired contents to make it easier for Internet users to visit desired websites. . They are also called portal sites because they act as gateways to Internet users on the Internet.

As portal sites grow exponentially, they are attracting more Internet users and new business opportunities. In other words, it is possible to make money by exposing advertisements of advertisers who want to promote their website to Internet users who visit the portal site.

In the early days of Internet advertising, banner exposure linked to advertiser's website was mainstream. However, with the evolution of the Internet, various internet advertising models were developed to increase the effectiveness of advertisements. Currently, portal sites are applying various new internet advertising models. .

The conventional technology of the newly appeared Internet advertisement model is applied to the 'computer communication network search engine' patented by Overture Services Inc. (hereinafter referred to as 'Overture Inc.') (Korean Application No. 10-2002-0045944). Location and price protection providing systems and methods of the list of search results generated by Overture's system provides a bidding tool that allows advertisers to have access to keyword search and registration tools and Internet users' search keywords to search for and register keywords associated with their website. The system provides a list of related keywords as a result of the search through the keyword search tool, allowing the advertiser to select related keywords and bidding desired search keywords through the keyword bidding tool. The system is set such that a list of websites can be arranged as a keyword search result of an internet user according to an advertiser's search keyword bid amount / requirement ranking for keywords bid from advertisers. The system is set to calculate the cost of advertising by clicking on hypertext, which means the visit of an Internet user, in the list of websites displayed according to the bid amount (hereinafter referred to as 'CPC'). have.

Another patent application of Overture (Korean Application No. 10-200--0069486), 'Location Bidding for Pay-Per-Search Systems by Location', suggests how to handle various situations according to advertisers' keyword bidding. That is, it includes content that allows advertisers who want the same bid amount and placement rank for the same keyword to manage the placement of the advertisement by raising the bid amount within a limit not exceeding the bid limit.

In search advertising systems such as Overture Inc., the method of efficiently selecting and recommending the best relevant keywords can have a significant impact on advertising, which is an essential and core issue. This is explained by its importance for the following reasons.

First, from an advertiser's point of view, if the relationship between keywords is not well known, most of the time, the intention to purchase only a specific keyword will be revealed, causing excessive bidding or purchase competition among advertisers for a few keywords. In addition, even if they find and select the relevant keywords by the intuition of the advertiser, the effect of the keyword selection based on the actual relevance is inevitably different. The generated advertisements inevitably include excessive competition price, and in some cases, purchase inappropriate keywords and execute advertisements, resulting in waste in terms of transaction inefficiency and cost effectiveness.

Second, in the case of a user who is a consumer of an advertisement, it may be an 'advertisement' or an 'information' depending on how appropriate the campaign is to meet the needs of the consumers due to the characteristics of the search advertisement. Another problem is that there is a high probability of being exposed to a missing advertisement. This means that even the medium providing the advertising platform suffers from the burden of inefficiency caused by indefinite (e.g., inappropriate keyword recommendation) in the process of supplying and demanding advertisements generated from the platform, thereby lowering profits. May result.

However, the above-mentioned advertising system of Overture Inc. is exposed to a somewhat inefficient aspect of keyword selection for advertisers to advertise their websites. It is important to search advertisers to recommend related keywords related to specific keywords to advertisers. Overture's system is only compound word level in providing related keywords. For example, it is limited to a function of recommending only a keyword including the word 'golf' to a customer (advertiser) who wants to purchase the keyword 'golf'. In other words, Overture's system can recommend a combination of keywords such as 'Golf Mania', 'Golf Course', and 'Golf Goods' to customers who purchase the keyword 'Golf', but 'Tiger Woods' and PGA that are highly related to 'Golf' Keywords such as' and 'skin games' are not recommended. This is a limitation of more relevant keyword recommendation in case of keyword recommendation system developed in other countries, not domestic companies.

In addition, Overture's system charges with a CPC method that calculates the cost per click of an Internet user. The CPC method has a fatal problem of 'click fraud' and may cause unnecessary expense to advertisers. In other words, the CPC method is charged according to the advertising effect, so unlike the existing exposure-based charging method, advertisers are currently using the most preferred charging method, but by intentionally repeating the click, it can promote the effect of the advertisement. there is a problem. This fraudulent click is a deadly risk that can undermine the CPC billing system, so many companies are building a system to prevent it, but it is difficult to prevent 100%. In addition, since click counting is difficult with web analytics, click counting in existing search engine search advertisements requires an intermediate server for counting clicks when a user clicks a search link. At this time, the intermediate server records each time one signal comes in, and counts the number of clicks through the aggregation. Ad platform for preventing click fraud caused by this process requires the cost of building a separate system. In addition, advertisers must pay for unintentional clicks by users or malicious clicks by the same person. There is a lot of inefficiency.

Meanwhile, in search advertisement systems using a conventional search engine, an advertisement is limited to a search list by a search engine. However, the expansion of advertisements through related keywords to various web documents will not only provide more advertising opportunities for advertisers but also provide higher advertising effects. In other words, portal sites not only provide search web pages to Internet users, but also provide various Internet services such as news, information, cafes, blogs, and the like. It is there. As such various web documents contain a large number of contents that match the search keywords of the Internet users, the visits of the Internet users occur very frequently. In addition, since a large number of keywords related to the interests of Internet users who visited the web documents are described, an advertisement related to the keywords described on the web document may have an excellent advertising effect. However, until now, advertisements using keywords in web documents have not been applied. Therefore, the necessity of advertisements using related keywords has been recognized in web documents.

An object of the present invention is to provide an internet advertising service system and method that can provide related keyword advertising that is effective in recommending and reasonable billing, and extended to web documents in order to solve the above problems and necessity. All.

In order to achieve the above object, the Internet advertisement service system of the present invention provides a search engine web server for arranging a list of websites having keywords related to a keyword search of an internet user in a predetermined form, and the search engine web server. A web analytics server that collects and analyzes keywords described in search keywords and web documents of an Internet user, and generates an association between the search keywords and the keywords described in each of the web documents; A database server for constructing a correlation between keywords generated in the web analytics server as a database, and recommending the keywords built in the database to an advertiser, and processing an advertisement according to bidding for the keyword desired by the advertiser. It includes an ad server.

The web analytics server generates an association for pairing each of the first keywords of the Internet user who uses the search engine web server and the search keywords after the first keyword, and the number of keyword pairs generated is the database. The server aggregates the search queries of all Internet users using the search engine web server and registers them in the database. The database server is set to be removed from the established database if the number of keyword pairs generated by the analytics server is less than a predetermined number for a predetermined period.

In addition, the web analytics server generates a pair association for each of the different words appearing in the web page document, the keyword pairs generated in this way by counting the frequency of occurrence for the web documents in the database server Register in the database. The database server is set to remove the association keyword pair having a frequency of occurrence less than a predetermined number of times for a predetermined period of time from the database.

The ad server provides a keyword search tool to search an advertiser related keyword list to advertisers who are willing to purchase at least one or more keywords, and the keyword server constructed in the database for a specific keyword input through the search tool. Provides a list of keywords that are associated with a particular keyword and pair.

In addition, the ad server presents a list of keywords associated with a pair of the specific keyword, and then each keyword along with the list of the keywords for the hyperlink selection means that the advertiser intends to bid on the keyword Provides a bidding tool for entering a bid amount for.

The bidding tool may be provided with information about bid amounts of advertisement competitors up to a predetermined rank for each of the associated keywords. In addition, the bidding tool may further be provided with data analyzed for the number of impressions and clicks for the associated keywords for each of the associated keywords. In addition, the bidding tool may calculate an estimate for the number of impressions and clicks for a predetermined period from the analysis data, and the estimated value may be further provided to the advertiser.

In addition, the ad server collects the key keywords in the website of the advertiser, and classifies the web document to the category classification for the advertiser according to the association between the collected key keywords and the web document Is set. The advertisement server applies a TFIDF (TermFrequencyInverseDocumentFrequency) weight to each of the keywords of the web document classified into categories, and gives bid priority to a website advertiser having the highest TFIDF weight.

The advertisement server may be configured not to issue a cookie having a predetermined session time in the session time, issue the cookie to an internet user terminal that has accessed the posted advertisement, and click the advertisement. Charge once based on cookies issued for. In addition, the advertisement server is set to be the payment processing to the advertiser in a pre-paid balance exhaustion method through the bidding account for the advertising amount.

The advertisement server is arranged to display advertisements on the web document while sorting advertisements according to the bid amount for keywords bid by an advertiser on a search result page of an Internet user. At this time, the advertisement displayed on the web document is set to the exposure ratio of the advertiser purchase amount to the total purchase amount for the bidding keywords. In addition, an advertisement exposed on the web document may be randomly exposed at predetermined exposure positions, or alternatively, may be set to be rolled and exposed at predetermined exposure positions.

The above internet advertising system can recommend the latest relevant keywords reflecting the trends of internet users when recommending keywords to advertisers, and can provide extended advertisements to web documents as well as per visitor rather than paying per click method. The payment method can provide a more reasonable billing system while eliminating the concern about fraudulent clicks, thereby enabling advertisers to have higher advertising effectiveness and efficient execution of the advertising budget.

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

1 is a diagram illustrating a network to which an Internet advertisement service system according to the present invention is applied. In the network, the Internet user / advertiser terminals 110, 120, 130 and the Internet advertisement service system 200 are organically connected through the Internet 150. Here, the Internet user / advertiser terminals 110, 120, and 130 may access the Internet advertisement service system through various Internet connection tools such as the desktop computer 110, the personal digital assistant (PDA) 120, the notebook 130, and the like. It is showing.

2 is a block diagram of an Internet advertising service system according to an embodiment of the present invention. The system includes a search engine web server 210, an analytics server 220, a database server 230, and an advertisement server 240.

The search engine web server 210 arranges and provides a list of websites having keywords related to a keyword search of an Internet user in a predetermined form.

The web analytics server 220 collects and analyzes keywords described in the web users' search keywords and the web documents provided by the system delivered through the search engine web server 210, and associates the web keywords with the relations of the search keywords. Create an association between the keywords described in each. To this end, the web analytics server 220 generates an association that pairs each of the first search keyword and subsequent search keywords of the Internet user who uses the search engine web server 210. In addition, the web analytics server 220 creates an association of pairs for different words appearing in the web page document.

The database server 230 builds an association between keywords generated by the analytics server 220 as a database. To this end, the database server 230 registers the association keyword of the pair generated by the analytics server 220 in the database, and targets a search query for all Internet users using the search engine web server 210. The number of generated keyword pairs is counted and registered in the database. The database server 230 is set to be removed from the established database when the number of keyword pairs generated through the analytics server 220 is less than the predetermined number for a predetermined period. In addition, the database server 230 aggregates the frequency of occurrence of the web documents with respect to the keyword pairs generated in the web document and registers the frequency in the database. The keyword is set to be removed from the database.

The ad server 240 recommends the keywords constructed in the database to the advertiser and processes the advertisement according to the bidding history of the keyword of the advertiser. And the charging is performed from the advertisement use of internet user. To this end, the ad server 240 includes a keyword recommendation processor 242, a bid management processor 244, a billing processor 246, and an advertisement processor 248.

The keyword recommendation processor 242 provides a keyword search tool to search a list of keywords related to ads to advertisers who are willing to purchase at least one keyword through membership registration, and provides a keyword search tool for a specific keyword input through the search tool. Provides a list of specific keywords built in the database and keywords associated with a specific keyword pair.

The bid management unit 244 provides a bidding tool for entering a bid amount for each keyword along with a list of search related keywords for hyperlink selection through an advertiser's search tool, which means a keyword bidding intention. The bidding tool includes information on bid amounts of ad competitors up to a predetermined rank for each of the keywords, information on the number of impressions and clicks for each of the associated keywords, and the number of impressions and clicks over a period of time calculated from the aggregated information. Estimate information for this, etc. are provided. On the other hand, the bid management unit 244 collects the key keywords in the advertiser's website, and classifies the web document to the category classification for advertisers according to the relationship between the collected key keywords and the web document. The bidding priority is given to a website advertiser having the highest TFIDF weighting value for each of the keywords of the categorized web document.

The billing unit 246 is configured to issue a cookie having a predetermined session time with respect to the posted advertisement and to be charged once based on the published cookie. The advertiser's settlement process is set up in such a way as to exhaust the balance through one-time charges in accordance with the issuance of cookies after the pre-charging of the advertising amount through the advertiser's bidding account.

The advertisement processing unit 248 processes the advertisements to be sorted according to the bid amount for the keywords bid by the advertiser on the search result page of the Internet user, and processes the position and the exposure amount of the advertisements exposed on the web document. At this time, the advertisement exposed on the web document is set to be exposed at the ratio of the advertiser purchase amount to the total purchase amount for the bidding keywords. In addition, the advertisement exposed on the web document is set to be randomly exposed to predetermined exposure positions. Alternatively, it may be set to be exposed while rolling to predetermined exposure positions. Of course, the advertisements displayed on the web document may be combined with a rolling and random method.

Hereinafter, a service processing process of the Internet advertisement service system according to the present invention will be described. In describing the service processing process of the system, the keywords described in the web documents are collected and analyzed to recommend keywords to advertisers and to display the ads on the web documents. Specifies.

3 and 4 are diagrams illustrating a keyword collection and promotion process through a search engine according to an embodiment of the present invention. First, when the search query is generated by the Internet user A searching the keyword 'golf' through the search engine web server 210, the search engine web server 210 transmits the search query to the web analytics server 220. Next, when User A searches for another keyword 'Tiger's, the analytics server 220 creates a relationship between the two keywords by pairing the relationship between the keyword' Golf 'and' Tiger's. do. The generated two related keywords are delivered to the database server 230 to build a database. In the same way, if Internet user B searches for 'golf' and then 'tailor maid', an association of 'golf-tailor maid' is created. Likewise, a pair of 'golf-golf goods' and 'golf-golf sky' is generated by user C and user D, respectively. As described above, the database server 230 counts the number of AB pairs and the number of pairs of keyword A and another keyword C in the same manner. You will create a set of associations between different keywords. On the other hand, there may be an irrelevant keyword pair in the established association set, the database server 230 removes from the database if the number of keyword pairs for a predetermined period or less. In addition, irrelevant keyword pairs can be manually removed. The association set of the keywords thus constructed is provided to the advertiser visiting the system 200 in the list as shown in FIG.

The result of the above work provides three advantages in advertising planning and execution.

The first advantage is that trends are continuously reflected in keyword recommendations. For example, when Tiger Woods gets married, his spouse, Elin Nordegrin, is also frequently searched for, and the association set is automatically updated to keep up with changing trends. This ensures that advertisers always receive the latest relevant keywords.

Secondly, even if you are a new golf equipment advertiser who doesn't have a lot of knowledge about golf, if you search for 'golf', a famous golf brand called 'Taylor Maid' will appear as a related keyword, so you can be more faithful to your website. But even if you don't know the details, you can achieve more scalability.

And the third most important advantage is that you can select the keywords that are best suited to you from the recommended related keywords. As shown in FIG. 4, there are various recommended keywords centering on the actual 'golf', but the related keywords presented have various characteristics such as golf equipment, simple interest, and game keywords for teenagers. In this case, the golf article seller can prevent unnecessary expenses by reviewing the proposed keyword and removing keywords that are far from their interests from the purchase list. For example, the online casual game, "You are the king of golf," will find that the associated keywords are mostly focused on the game, and thus can be removed from the purchase list. In short, among relevant keywords, you can filter out keywords that are appropriate for you.

Meanwhile, the Internet advertisement service system of the present invention includes an additional keyword collection process as well as an advertisement through a search engine as described above. 5 and 6 illustrate a keyword collection and promotion process through web documents according to an embodiment of the present invention. The analytics server 220 and the database server 230 generate a pair between keywords appearing in one web document and count the number of pairs for the entire web document. For example, if different keywords a, b, and c are listed in the sports news document A provided by the online media 'Media Daum Server 510' provided by the portal site 'Daum', the keywords a, b, By combining c into ab, ac, and bc pairs, a keyword set is constructed for one keyword as in the result of the example shown in FIG. Similarly, a plurality of associated keyword pairs can be obtained from a plurality of words described for Document B. With respect to the association keyword pairs thus obtained, the database server 230 aggregates the number of pairs of the entire web documents into a database and recommends the association keyword as shown in FIG. 6 at the request of the advertiser. In addition, keyword extraction for web documents is also performed for web documents of cafe server 520 and other document server 530.

As described above, the extracted keywords in the web document are applied to the advertisement through the search engine so that the latest keywords can be updated more quickly. Meanwhile, the web document extraction keywords may be provided exclusively to advertisers who want to advertise on the web document page. To do this, we build keyword pairs in web documents into a separate database, provide advertisers with a selection tool to choose between ads using search engines and ads using web documents, and then recommend keywords based on the advertising method selected by the advertiser. Can be performed.

In the web document association keyword pair extraction as described above, the keyword set generated by the search engine may partially cover the keyword in the document for the advertisement of the advertiser who purchased the keyword for the web document. The keyword set generated by searching in non- or colloquial words is also to be considered when the written word does not match the written word and the exact standard word. In other words, if keyword recommendation using a search engine is due to user behavior, document-oriented keyword recommendation may be attributed to the frequency of keywords frequently appearing in a document at the time of document generation (news supply, personal creator). The latter must be added by the nature of the web document advertisement that finds an advertisement suitable for the nature of the document.

In addition, the keyword extraction for keyword recommendation in the system of the present invention reflects both methods of extracting an associated keyword using a search engine and extracting an associated keyword through a web document for a search keyword, and even a compound word including a search word therein. It is set to provide a list of related keywords. Accordingly, the system can recommend an optimal keyword reflecting the trend to the advertiser.

The advertiser can then place his ad in a predetermined location by participating in the bid for the recommended keywords. 7 and 8 illustrate a keyword bidding process according to an embodiment of the present invention. First, the advertisement processing unit 248 obtains click information (2) by the user A clicking (1) the advertisement exposed in the web document, and delivers the exposure information and the click information to the bid management processing unit 244 (3). . The bid management processor 244 provides the advertisement exposure information and the click information delivered to the advertiser with the list of recommended keywords, and provides a bidding tool for filling in the bid amount (4). When the advertiser enters the bid amount for the keyword through the bidding tool and confirms the bid intention (5), the bid processing unit 244 delivers the advertisement placement information according to the keyword bid amount to the ad processing unit 248 (6). The advertisement processing unit 248 exposes the advertisement 7 to the search list of the web document or the search engine according to the received advertisement placement information.

Here, the contextual advertising has a different key difference from that of the existing search engine. In general search engine advertising, the number of impressions (search queries), clicks, and bid amount of keywords are important factors. In the case of document advertisements, advertisements that match web documents may have higher advertising effects. Therefore, it is important to select keywords that are most suitable for the characteristics of documents among numerous keywords existing in web documents. Exposure is an important factor. However, while a search engine advertisement is a 1: 1 matching that exposes an advertisement of an advertiser B who has purchased keyword a to a search for keyword a of user A, as shown in FIG. Since N keywords among the keywords can be referred to as N: 1 matchings among advertisers, the advertisement exposed in the web document is for each advertiser who purchased N keywords in the document as shown at the bottom of FIG. Advertisers are set to adjust the amount of advertising exposure according to the bid amount.

In addition, as mentioned above, in addition to selecting which keyword is most suitable for the nature of the document, the advertisement most suitable for the document will have a great influence on the advertising effect. Accordingly, the bid management processing unit 244 classifies the web document into a plurality of categories, and then expands the bidding priority to advertisers matching the category. The category applies a category classification that categorizes advertisers. To this end, we collect key keywords from advertisers' websites, analyze the association between collected key keywords and web documents, and classify web documents according to the associations. Keywords in the categorized web documents are assigned a TFIDF (Term Frequency Inverse Document Frequency) weighting, and bidding priority is given to the website advertisers with the highest weights. In addition, the TF-IDF algorithm is most often used to distinguish which document is more related to a particular word and further determine the priority of the document when the word is related to the specific word in the document. The term frequency (TF) indicates the number of times a word appears in a document, and the document frequency (DF) indicates how many documents a word appears in a set of N documents searched for. If a certain search word appears in a document, the document can be judged as an important document for the search word, but if it appears in several documents, the importance of the word is less. Therefore, the priority of the document is expressed by multiplying the TF value by the inverse document frequency (IDF) value of the DF.

 On the other hand, in the bidding process as described above, by providing additional secondary information to the bidding tool can give a greater merit to the advertiser. In other words, the direct driver for advertisers to temporarily select the desired keywords and then actually initiate purchases and bidding activities is secondary information such as competitors' bid amounts, impressions, clicks, and estimates. It is often based on. Advertisement budgets are pre-charged and paid on clicks for ads and contact ads using search engines. For advertisers, the number of impressions and clicks for certain keywords in the past, as well as fluctuations in bidding amounts of existing competitors You will be able to set up your own advertising campaign. It can be said that purchase and bidding are largely determined by how faithfully and reliably the advertisers provide the materials they can refer to when purchasing. Accordingly, the system of the present invention provides a bidding tool including various secondary information in order to faithfully deliver bidding related information to an advertiser. 9 is a view showing a bidding tool according to an embodiment of the present invention.

Next, when the bidding process is completed through the bidding tool, the system exposes advertisements according to the bid amount and / or priority applied by advertisers for the keywords. 10 and 11 illustrate a process of displaying a contextual advertisement according to an embodiment of the present invention. When the advertisers A, B, C, and D bid on the contextual advertisement, the bid management processor 244 first calculates an exposure amount according to the bid amount of each advertiser, and notifies the advertisement processor 248 of the exposure amount. Then, the advertisement processor 248 exposes the advertisements of the advertisers on the web document according to the exposure amount information of each advertisement delivered from the bid management processor 244. At this time, the advertisements exposed on the web document are set to be exposed at the ratio of the advertiser purchase amount to the total purchase amount (total bid amount) for the bidding keywords.

In general, people's eyes tend to move from top to bottom and from left to right when accessing web documents. This is especially true for web pages that are the result of search engines for information acquisition purposes. Therefore, it is known that the most effective at the top, in the order nearest to the search box. Therefore, it is common for the advertiser who presents the highest amount of bidding results to occupy the highest position. However, in the case of contextual advertising for non-search webpages, the characteristics are somewhat different. Many documents on the Web come in different forms. In the case of the Internet media, 'Daum', the bottom part of the article is most noticeable, and the right end, which is hardly noticeable in general web documents, is also recognized as a spot where it is attracting attention. It is becoming. In other words, depending on how the document is edited, the user's attention is different. Therefore, the approach should be different from the search engine advertisement that occupies the top position at a high price, which is shown in the uniform and fragmentary user interface provided by some search engines. will be. That is, in the editing as shown in FIG. 11, the area of the area 2 is expected to be the most effective in the zones 1 to 4, but in order to prove this, it is not necessary to calculate and apply the advertisement effect verification work for many editing forms. Therefore, adjusting the exposure amount by rolling and / or randomly exposing various areas may be the most reasonable for the advertisement on the web document.

Therefore, the system of the present invention is set to adjust the amount of exposure according to the amount of advertising, rather than determine the position of the advertisement according to the amount of the web document. This method will be the most advantageous and effective advertising exposure method for advertisers who presented a high amount regardless of the user interface.

Meanwhile, after notifying the advertisers in advance for efficient operation of the system and maximum exposure of the advertisers, the exposure may be excluded from the advertiser advertisement having a predetermined lower purchase amount compared to the total purchase amount. For example, the amount of exposure may be set so that the remaining 90% of the advertiser is excluded from the advertisement of the small advertiser belonging to the lower 10% of the total purchase amount of the keyword.

In addition, the Internet advertisement service system of the present invention applies a Cost Per Visitor (CPV) method to eliminate the waste of the budget caused by duplicate clicks on the same person as the risk of fraudulent clicks. In this method, when the user clicks, the ad server 240 issues a cookie having a predetermined session time to the user's terminal, and counts the cookies issued without issuing a cookie within the session time. Charge the bill. In other words, even if a large number of clicks are generated from a terminal within a short time, one visitor is regarded as a cause and charges for one cookie. 12 is a view illustrating a charging process according to an embodiment of the present invention. When the user clicks on Advertiser A of Advertiser b, the advertisement processing unit 248 first checks whether the user's client has a cookie older than 3 minutes before issuing the cookie. If there are cookies older than 3 minutes, delete them. 2) If the cookie is within 3 minutes, check whether there is a cookie of the same advertisement A. 3) If there is a cookie of the same advertisement A (because it occurred by clicking within 3 minutes), it is charged and only the cookie last time is updated. Charges if Ad A's cookie is missing.

Even if User A generates several clicks within 3 minutes from his laptop, the ad server issues only one cookie, and only counts the cookie so that only one charge occurs. How long the session takes is a matter of advertising platform operation. As another example, while User A clicks on User A's notebook several times within three minutes, even if User B clicks on User A's notebook, the cookie is already issued, so only one charge is made. This does not cause duplication of charges in the case of the user C in the case of using the user A's terminal. However, in the previous two examples, if a click occurs after the session time, a cookie is generated once again and billing is performed. This not only solves the problem of click fraud, but also has the advantage that the advertiser does not have to pay for unnecessary clicks.

In FIG. 12, a billing method regarding a plurality of clicks for one advertisement generated within a specific time in one client (computer terminal) has been described. As another example, when a plurality of different advertisements are clicked on one client The following describes the charging method.

The user's browser limits the cookie storage to 255 bytes for a single DNS. Therefore, when the user clicks on the number of advertisements exceeding the cookie storage capacity within 3 minutes, the browser may continuously update the cookie information, thereby charging the same advertisement. For example, if a user clicks on advertisement A, a cookie is issued, then clicks on dozens of advertisements other than advertisement A within three minutes, removes advertisement A from the user's cookie storage, and then the cookie is issued and billed again. In this case, by checking the time with the session ID, clicks within 3 minutes can be again classified, and whether or not to charge or not charged. Such a method can be performed smoothly unless it is generally abusing in a way that is faster and cheaper than when determining a cookie in a web server. In addition, abusing clicks can be prevented by identifying the second session with the secondary session ID.

In addition, the Internet advertising service system of the present invention performs a variety of advertising reporting to support the advertising policy of the advertiser. For example, a web page for reporting a click rate / impression amount / competition / exhaustion amount by section, keyword, and date by a keyword purchased by an advertiser is provided.

As described above, the Internet advertising system of the present invention can remove unnecessary bubble cost in both the advertiser and the advertising platform, so that the former enables more efficient execution of the advertising budget and the latter is more economical. To increase profits.

In addition, the system of the present invention supports the advertiser to reflect the latest trends to the advertiser through a keyword that is constantly updated.

In addition, since it is possible to expand and apply advertisements using related keywords to various web documents, not only can provide advertisers with more advertising opportunities but also support maximum advertisement effect.

Claims (40)

A search engine web server for arranging and providing a list of websites having keywords related to a keyword search of an Internet user in a predetermined form; A web that collects and analyzes keywords described in web documents and search keywords of an Internet user delivered through the search engine web server, and generates an association between the search keywords and the keywords described in the web documents Log analysis server; A database server for constructing a relation between keywords generated in the web analytics server as a database; And And an advertisement server for recommending the keywords constructed in the database to an advertiser and processing an advertisement according to bidding for the keyword desired by the advertiser. The method of claim 1, The web analytics server generates an association relationship between each of the first search keyword of the Internet user and the search keywords after the first search keyword as a pair and delivers the correlation to the database for registration in the database. Internet advertising service system characterized in that. The method of claim 2, And the database server aggregates the number of association keyword pairs as a search query of all Internet users and registers them in the database. The method of claim 3, wherein And the database server removes from the database when the number of association keyword pairs generated by the analytics server is less than a predetermined number for a predetermined period. The method of claim 1, And the web analytics server generates an association of pairs for different words appearing in a web page document and delivers them to the database server for registration in the database. The method of claim 5, The database server aggregates the frequency of occurrence of the association keyword pair appearing in the web page document to the entire web document and registers in the database. The method of claim 6, And the database server removes the association keyword pair having a frequency of occurrence less than or equal to a predetermined number of times for a predetermined period of time in the web documents. The method of claim 1, The ad server provides a keyword search tool to search an advertiser related keyword list to advertisers who are willing to purchase at least one or more keywords, and the keyword server constructed in the database for a specific keyword input through the search tool. Internet advertising service system, characterized in that providing a list of keywords associated with a particular keyword and pair. The method of claim 8, The advertisement server presents a list of keywords associated with the specific keyword in pairs, and the advertiser selects a list of the keywords with respect to the hyperlink selection that means the bidding of the keyword. An internet advertising service system comprising: providing a bidding tool capable of filling in a bid amount. The method of claim 9, The bidding tool Internet advertisement service system, characterized in that for each of the associated keywords to provide information about the bid amount up to a predetermined rank of the ad competitors. The method of claim 9, The bidding tool Internet advertisement service system, characterized in that for each of the associated keywords further provides data analyzed for the number of impressions and clicks on the associated keywords. The method of claim 11, The bidding tool calculates an estimate for the number of impressions and clicks for a predetermined period of time from the analysis data, and the estimated value is further provided to the advertiser, the Internet advertising service system. The method of claim 1, The ad server collects the key keywords in the advertiser's website, and classifies the web document by applying to the category classification for the advertiser according to the association between the collected key keywords and the web document Internet advertising service system. The method of claim 13, And the advertisement server applies TFIDF (TermFrequencyInverseDocumentFrequency) weighting to keywords of the web document classified into categories, and assigns a bid priority to a website advertiser having the highest TFIDF weighting. The method of claim 1, The ad server is configured not to issue a cookie having a predetermined session time within the session time, issue the cookie to an Internet user terminal that has accessed the posted advertisement, and click on the advertisement. Internet advertising service system characterized in that the charging once based on the issued cookie. The method of claim 15, The ad server is set to pay the advertisers in the precharge balance method for the amount of advertising, the Internet advertising service system. The method of claim 1, The ad server sorts the advertisements according to the bid amount for the keywords bid by the advertiser on the search result page of the internet user, and processes the advertisements to be exposed on the web document for the web document selection of the internet user. Internet advertising service system characterized in that. The method of claim 17, The advertisement displayed on the web document is set to be exposed at the ratio of the advertiser purchase amount to the total purchase amount for the bidding keywords. The method of claim 18, The advertisement exposed on the web document is set to be randomly exposed to the predetermined exposure position Internet advertising service system. The method of claim 18, The advertisement exposed on the web document is set to be rolled and exposed to predetermined exposure positions. Collecting keywords described in search keywords and web documents of an Internet user delivered through a search engine web server; Generating an association between the search keywords and the keywords described in the web documents; Establishing an association between the keywords in a database; Recommending the keywords built in the database to an advertiser accessing the system; Providing a bidding tool configured to fill in an amount together with bid related information for the keywords recommended to the advertiser; And And displaying an advertisement according to the bidding contents for the keywords set through the bidding tool. The method of claim 21, The generating of the correlation between the search keywords may include generating a correlation between a first search keyword of the Internet user and each of the search keywords after the first search keyword in a pair. 23. The method of claim 22, wherein the database building step Aggregating the number of the keyword pairs whose associations have been generated for the search keywords to a search query of the entire Internet user; And And matching the number of the keyword pairs, which are counted for a search query of the entire Internet user, to each of the keyword pairs and registering them in the database. The method of claim 23, wherein The database building step may further include deleting the generated keyword pairs from the database if the generated keyword pairs are counted less than a predetermined number of times for a predetermined period. The method of claim 21, Generating an association between the keywords described in the web documents, and generating pair associations for different words appearing in each web document. The method of claim 25, wherein the database is Counting the frequency of occurrence of the association keyword pair appearing in the web document for all the web documents; And Registering the frequency of occurrence of the association keyword pairs aggregated for all the web documents in the database. The method of claim 26, And if the database is less than a predetermined number of times for a predetermined period of time for the association keyword pairs generated from the web documents, deleting the data from the database. 22. The method of claim 21, wherein recommending the keywords to advertisers that have accessed the system Providing a keyword search tool to search an advertiser-related keyword list for advertisers who are willing to purchase at least one keyword; And And providing a list of keywords related to the specific keywords and pairs established in the database with respect to the specific keywords input through the search tool. The method of claim 28, The providing of the bidding tool may include providing a list of keywords associated with the specific keyword in pairs, and then selecting each of the keywords along with the list of the keywords for hyperlink selection that means the advertiser intends to bid on the keyword. An internet advertising service method, comprising: providing a bidding tool for entering a bid amount for a keyword. The method of claim 29, The bidding tool is provided with information on the bid amount of each of the associated keywords until the transfer of advertising competitors up to a predetermined rank. The method of claim 29, The bidding tool is an Internet advertising service method characterized in that the data analyzed for the number of impressions and clicks on the associated keywords is further provided for each of the associated keywords. The method of claim 31, wherein The bidding tool calculates an estimate for the number of impressions and clicks for a predetermined period from the analysis data, and the estimated value is further provided to the advertiser, characterized in that the Internet advertising service method. The method of claim 21, wherein providing the bidding tool Collecting key keywords from the advertisers' websites; And And classifying the web document by applying a categorization to the advertisers according to the association between the collected core keywords and the web document. The method of claim 33, In the step of providing the bidding tool, the Internet applies a term prioritization weighting value (TFIDF) to keywords of the web document classified into categories, and gives a bid priority to a website advertiser having the highest TFIDF weight. Advertising service method. The method of claim 21, Setting a cookie having a predetermined session time not to be issued repeatedly within the session time; Issuing the cookie to an internet user terminal accessing the posted advertisement; And And charging once based on the cookie issued for the click on the advertisement. The method of claim 35, wherein The billing is Internet advertising service method characterized in that the advertiser is set up to be charged in a precharge balance. The method of claim 21, In the displaying of the advertisement, the advertisement is arranged on the search result page of the internet user according to the bid amount of the keywords bid by the advertiser, and the advertisement is displayed on the web document for the selection of the web document of the internet user. Internet advertising service method characterized in that. The method of claim 37, wherein The advertisement displayed on the web document is set to be exposed at the ratio of the advertiser purchase amount to the total purchase amount for the bidding keywords. The method of claim 38, The advertisement exposed on the web document is set to be randomly exposed to a predetermined exposure position Internet advertising service method. The method of claim 38, The advertisement exposed on the web document is configured to be rolled and exposed to predetermined exposure positions.

Images