TW200832268A - Category-based advertising system and method - Google Patents

Abstract

A category-based advertising system is disclosed. An analysis module extracts one or more web documents including a main keyword from a plurality of web documents. A keyword extraction module extracts one or more words included in the extracted web documents. A weight calculation module calculates the weight of each of the extracted words with respect to a corresponding category based on the exposure characteristics of the word on the web documents including the main keyword of the corresponding category. A category matching module sets one or more words as associated keywords, and generates one or more category sets with which associated keywords and weights thereof are associated. An advertisement matching module determines the category of each of the corresponding web documents using words included in the corresponding, web document and the category sets, and matches the advertisement of an advertiser to the corresponding web documents.

Description

200832268 IX. DESCRIPTION OF THE INVENTION: TECHNICAL FIELD OF THE INVENTION The present invention relates to an advertisement system and method based on categories, and more specifically, Text, advertisers are interested in this. [Prior Art] ^Shi Xi ^月升乂" can provide, for example, news items, blogs and search for "web documents"), and documents include advertisements.广罔,文^中' usually corresponds to multiple keywords, which are included in the web page file and displayed as needed. Yuan and 2 = 2 informants register 'character' bag (four)), and according to the corresponding word Du H) 'Advertiser's advertising combined with characters, bags, and nets such as dedicated posts web materials, riding In this method, 'because the advertisement meets a specific character, the resulting = title is in many cases unable to perform the matching of the # advertisement. In: 兀''hand (four) (handbag), included in the _ webpage The document financial element, the generation ^, is not included in the case of this web page file, the registration character "bag two children and does not combine with the corresponding web page file. In addition, in the case where the character registered in the advertisement surface is included in the web page but the overall text of the corresponding web page file is inappropriate, the text in the text is not combined in the text. For example, if the advertisement is included in the news story "the murderer's body is lost in the child" 51, the advertiser will not be provided with the news story irrelevant = better 6 200832268

However, when the advertisement is executed only by the matching character, as in the prior art, in order to solve the problem, the Korean Unexamined Patent Publication No. 10-2005-0058172 discloses the advertisement system and method, which uses the advertisement of the special person. Classify and combine with the corresponding text. 2 keywords so that the advertisement and the text are not combined with each other, wherein the text includes a proprietary keyword with negative meaning. In the specification of Korean Unexamined Patent Publication No. 1〇-2〇〇5-〇〇58172, the above problem can be solved by preventing an advertisement of an advertiser from being combined with a text having a proprietary keyword. In addition, in the Korean unexamined _ check patent publication number No. 1〇-2〇〇5.8172, the text and advertising

For example, a news story "a body of a murderer is thrown into a bag" is included in the category, social/cultural/news, and the advertiser's advertisement is included in the category "shopping/fashion/bag," , in order to prevent the advertisement from being combined with the news story, because there is no connection between the news story and the advertisement. Here, the category is set by the advertising broker and hierarchically (hier muscle hically), and the advertiser selects One of the plurality of categories, of course, has a plurality of webpage files that preferably conform to the advertisement of the advertiser, even if the webpage file is included in a face towel different from the face of the borrower (four). 'Korean Unexamined Patent Publication No. I child month ^ in the issue of advertising can not be combined with these web documents, ΪΓΙ ΪΓΙ, 倾 or category "social / cultural / film,, the situation is

The advertiser's ad will be combined with the web page file. Of course, South Korea is not in the category selected by the advertiser, shopping / fashion / bag, and web file package 7 200832268 = check the open 5 tiger brother! (4) Still rhyme 5 8! No. 72 manual The advertisement is combined with these webpage documents, because the categories of webpage documents may differ. "The category and advertisement 1 selected by the person, according to the method of shooting, [invention] Therefore, the present invention is intended to solve the problems in the prior art - the purpose is to provide - _ class based on n pure and ^ used to analyze the website The documents provided (or the internal text of the eight ',,, the law ' ^, or the off-boat read, the file package domain 2, the other two, Li Tong, two other uses to provide a category based on: two related key The word 'and the connection to find a ^ system of its reuse one to = set in ^ mention = kind of lang based on the choice of advertising - _, _ advertising exposure; ^ read child % 'money advertiser can (four) this: Another purpose of the present invention is to provide a wide-ranging basis based on categories. The system is a page file, and the web page file package is extracted from one or more cyberspaces. Include a primary keyword from the number of web pages. Μ -. 避 2 group extracts one or more words included in the extracted web file = ... material module relative to the corresponding category based on the character re-web file Calculate the weight of each extracted character by exposing the eXp_e) property, web page 8 2008322 68 words. Category matching module settings - or multiple characters are: Guan 2: 'Sub-related key words and weights to generate one or more characters to determine the category of each-phase_page file, and = web page file matching - σ 者 的 的 的 的 的 的 的 的 的 的 的 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The key (4) corresponds to the curry 4 'based on the character (four) light features relative to the weight of the other main key elements 'web file includes the corresponding class set the target number or the eve of the child to make one in each corresponding category The pottery rights should be based on the plural related keywords of the category, and based on the relative keywords and classes ==^ or multiple categories set t and use the included in the web page file and other sub- 7" to determine the category of each-web file, and An advertisement registered as a class scatterer matches the corresponding webpage file. Exposure ίί=ΐϊ—Average exposure frequency and—Exposure concentration, average usage: The frequency of exposure in each file, exposure concentration ίί Number of documents and text The ratio of the total number of "-" is different. "When calculating the exposure degree, the weight calculation module specifies: the purer weight is the - character included in the -web file, and the webpage file is extracted f Station (sling relative to each category, the station has a higher weight. In addition, when counting the exposure of the towel = material, the weight calculation of the ember - subtraction of the weight in the character = position of each position - word 疋. In addition... The position of the character is - the title part of the file I, the body „卩分和—the background material part of the ―, and the weight of the position according to the title part, the body part and the background material part*. 9 200832268 Compliance with a pre- Each relevant keyword of the criterion includes a weight having a level equal to or higher than a preset level of the corresponding category settings. Moreover, preferably, the specific character that exists only in one category is a related keyword and has the highest weight with respect to the corresponding category. In the present invention, the category setting can be added periodically or irregularly. According to the present invention, when a file of a particular category is extracted, a file that does not include the main key but includes a corresponding category of the related keyword can be taken in accordance with a preset reference. That is, each of the two files including the related keywords will be classified according to the preset order of the weights, each file can be extracted, and each file includes a special sub-element, k is consistently present in a category. Specific characters. [Embodiment] - The following reference is made to the same or similar elements or devices in the different illustrations. As the = diagram axis shows the miscellaneous 剌 阙 类别 类别 ( (4) I Yin, the method of notifying, this figure shows the category and phase keywords

Good body, _, 蝴,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, —— The phase _ key word in the category, this | two table includes the same name in the body. The key is ready-to-eat, the primary keyword and the _ other advertiser refer to the plural category and the categories included in the advertisements that are included in each -_ key selection. 200832268 '=This = in the category-based advertising method, using the category δ and then unloading the class - or multiple web pages for the special recognition" is included in at least one character of the relevant document, and the phase is at 4 ^ The award will be combined with the corresponding documents based on the notice. When the network uses a wide range of other advertisers, the advertisement is a document that is not combined with the advertisement according to the present invention, and will be periodically or irregularly = for example, when each recent or periodic registration When the webpage file is a new webpage to be analyzed, the self-contained library network Baidudu, the child lies in the Λk (4) key, and the base (4) should catch and search the brother and the existing related keywords. Therefore, the use of the analytic method to continuously receive the extra-collection is based on the fact that since the complex digital element corresponds to the borrower key, the character is recently used in a particular category, so; == light opportunity thus increasing the advertising effectiveness. Another * ^ wide mouth exposure = character 赖 ’ ‘users can make coffee; make == the content of the article has a South relevance of the advertisement. Figure 3 is a block diagram of a notification system in accordance with the present invention. According to the present invention, the category is used as the base module 11G, and the keyword extraction module 5 is wide-^T includes one-to-one _pi_&14〇, 1#|= The counting module 130, the analysis module UG fine = Γ and - the advertising is short (10). Page files include files from multiple web pages \=page files 'mother one net' For example, in 镰_,, dieting! keywords. When you enter the homework analysis, including the character, the section 200832268, each of the webpage files will be extracted from the default website, and the keywords of the category should be included. The sub-elements represent the pair and can further exclude the main_file. Relevant keywords are based on the weight of the single key related to the extraction of L-files outside the 'specific keywords between the relevant keywords are included

Considering each category, the keyword extraction module (10) uses a morpheme analysis method for the web page file to be taken, and - or a plurality of characters in the Jiang Guo 0. The sub-το-to-weight calculation module 13〇 is provided by the keyword _3 & ~120 according to the corresponding category and reference to the degree of light extracted from the file and the number of files corresponding to the exposure of the character; ^ exposure rate, exposure key secret module 12 〇 extracted each block (10) calculated borrow weight 7 The number of files including the exposure concentration calculation module 131, the station weight center group 132, the reliability calculation module 133, the group 131 light t: the weight calculation module 135 and the second weight calculation module And calculate the per-exposure forest yuan (weight. Riga 戚 Hi a screening) when the number of documents exposed by the relevant characters is reported less, then increase:; Cheng Wei: This is because the characters are only in the special exposure The number of files I is very small, but the frequency of the characters in the exposure file is more than the number of exposed mosquitoes and the frequency of the relative file (four) optical characters ^ = the corresponding word is - for example, In the category "dental clothing spears" for example, doctors, ',,, hospitals, and, nurses, the characters will be frequently exposed and the characters are exposed to 12 200832268 light,! Niu, Li Xi, make this The word can be judged as the general word of the medical field order. Conversely, _ has a mother s shirt, and moves : Relatively few 'These words are frequently used in categories, dental services, exposures, 丄,,, dental services, and related keys, these words have a higher weight. Just made, say!/read complicated The document of the exposure amount of the towel collection is said to be Λ Λ ° is assigned to the ratio of the weight of the material, and the value of the health increases with the increase of the plant degree. The weight can be calculated by using, for example, the following formula. N log(TF/(iDF + 1)), where the average exposure frequency of the file is 'Ν', the number of files is 4, and the number of files with the given character is used. The first -_ and the sub-group (4) are analyzed and set as U), 〇〇〇, 〇〇〇, in a total of 10,000, _ pieces of documents,,, ^ body f exists in 10, _ files, system This number will be analyzed, and in each file the character "funnel shape", the average appearance of 5 times, the weight of the figure '' will be calculated as follows: /, 5: 1 1 (10,00 〇, 〇〇〇 / (i〇, 00〇+1)) and 2 49 Assume that the number of articles analyzed according to the first category and analyzed by the analysis module 10 is 1〇, 〇〇〇, 〇〇〇, in the division The total number is 1〇, _, _ piece Feng 2, the character “weight” appears in the 3,000,000 construction file, and the character “weight”, the average number of occurrences in each file is 10 times, the character, the weight, the weight will The calculation is as follows: 10 10 log (10,000,000 / (3,000,000 + 1)) 230 As described above, the weight calculation method according to the present invention does not assign a high weight to - the character 'only because the character has a high exposure rate in the complex shot. According to the present invention 13 200832268: Re-concentration first exposed in a limited number of file characters in a class with a higher station weight calculation module, quite 132 based on the corresponding? Yuan's file source (website) and increase f is less per-character (four). For this purpose, the weight of each website is calculated based on the rating of the webpage file between the webpages of the webpages in the specific webpage based on the rating of each webpage. According to this, according to the per-character, the station weight is calculated by combining the weight of the website with the station weight of the website, and the website includes a web page file having the corresponding character.

The light collar 2 heavy X knot f straw increases/decreases the exposure of the exposure characters in a particular website. For example, in the calculation of categories, dieting, and the weight of each-character weight in the month of 'Yu Zhanzhong 皞# Xuan $+ 适,上#丄,1 *先子兀的第一先The degree of concentration can be used, for example, by increasing the number of categories according to the category, and for dieting, the station right of station A is greater than the preset value. That is, if the character is a good figure, it will appear in the station A _ times, and the gambling of the gambling will be counted as 15 (): the weight of the station: according to the character appearing in the main station Calculated by the quantity, and the second = upper level ^^ is calculated by calculating the charm 131 by the exposure concentration degree. The positional weight calculation module 134 assigns different weights to the position - the word is based on the collected character body part and the background material part, and the position weight is fine, = part = the younger brother 6 picture is based on the file An illustration of an example of a word. On the first day, the weights are different to one

Na-Ycnmg Kang" is written in the title: The main purpose of the magical cow in the 4 inch children is to be publicly displayed on the subject of 200832268. The model of the wounded 61''1^_¥〇11叩1^ tons, And the diet is an affiliate information. The location weight metering module 134 assigns the highest weight to the character, where the position of the wood in the character corresponds to the header portion, and the location weight calculation module assigns the weight of the person n to the body portion. And assigning the lowest weight to the corresponding word of the background material = bit. When calculating the frequency of the corresponding character, it can reflect such weight, that is, the character that appears in the title part of the rice sip is exposed twice and out. Now the ^1* 夂 of the body part can count the exposure frequency of the corresponding character. In addition, the weight can be calculated based on the degree of the word= appearing in the title part, and multiplied by a weight, which is concentrated by the exposure concentration The trust weight calculation module 133 changes the weight calculated by the exposure concentration meter #module 131 according to the reliability of the station. The reliability of the station can be confirmed by the preset field (for example, ''dieting') Corresponding to stand for a preset period Judging by continuous management. The single weight is found and collected from two or more categories, and the first weight calculation module 135 and the second weight calculation module 136 perform a judgment operation on the category in which the collected characters are located. Figure 4 is a schematic diagram conceptually showing a category-based advertising method in accordance with an embodiment of the present invention showing two or more categories consistent with a single character. Please refer to Figure 4, the characters shown in Figure 4, scale, can be exposed to categories, diet, (here, referred to as the first category) for description and understanding) and category "sports equipment" (herein referred to as the second category, for explanation and understanding). If the word ''proportion' is used as the keyword 'character' extracted from the keyword extraction module 120 according to the category "diet" 'also exists in the category "sports device", the weight calculation module 130 sets the category "diet" as the first category and sets the category, and the sports device" as the second category. The first weight calculation module 135 calculates the word The weight of the first category, and the second weight calculation module 136 calculates the weight of the second category of the word "20088268" "sports device", and the calculation result is transmitted to the category matching module 140, such as Figure 5 shows. Optionally, the weights of the characters having two or more categories may be calculated according to the relative categories, and the calculated weights may be compared with each other, and therefore, the category having the larger weight is selected as the category of the corresponding character, and The selection result will be transmitted to the category matching module 140. Alternatively, a single character may be included in the complex digital element settings, since the corresponding characters may differ according to different categories. For example, in category, dieting, the weight of the character "proportion" can be 5.2, and in the category, sports equipment,, the weight of the character "proportion" can be 3.1. Category matching module 140 generating a category setting, in which the character matching the preset criterion in the corresponding category is related to the weight of the category setting according to each category. The category matching module 140 may include the related keyword determining module 14ι and the category setting. The module 142 is generated. The related keyword determining module 141 sets a plurality of characters, each of which has a weight equal to or greater than a preset reference value, or each of the characters has a weight, and the weight has a level, and the level is equal. Or greater than the preset level. According to each category and referring to the weight of the character providing the self-weighting module 130, the character is regarded as a related keyword of the corresponding category. For example, a plural method, such as setting a character, may be employed. Method, each character having a weight equal to or greater than 3 can be regarded as a related keyword of the corresponding category; setting a method with the highest weight of 20 characters as a related key of the corresponding category' And setting a method having a related keyword greater than 2 and the top 20 high-weight character as a corresponding category, etc. - the category setting generation module 142 generates a category setting based on having a high-weight related keyword in the corresponding category, and The storage category is set in the database. The category is set to the related keywords and their weights are related to each other. In the relevant key 200832268 words included in the category 'dieting' is "good body",, funnel-shaped body, and,, obesity , and in the case where the weights are 9, 8 and 7, respectively, the category setting indicates the character, the good figure ', the combined weight 9, the character "funnel shape", the combined weight 8 and the character, obesity, In the absence of information on human rights, w" At the same time, in the case where the primary key is used and the character is found to be only found in a particular character of the corresponding category, the category matching module 14 can assign the highest weight to the specific character. In the case of high specific projects, the project only appears in a specific category, = it has an exposure frequency. Therefore, although the corresponding character can indicate a specific category, its warm light frequency is not high. Therefore, since the character has a low weight, the word = the relevant keyword of the corresponding category may not be registered. ^ To prevent this situation, the category matching module is early red - 70 °, for example, 'infrequently used characters, pulmonary fibrosis And, in the web file including this character belongs to the category "lung, even according to the calculation (four) method, character, job dimension, the low value of the right, the judgment character, the lung fibrosis, - for For the specific word S, the word 2 = dimensioning is set to have the highest weight. It is judged that the exposure frequency in a specific word is almost. But =^2: ## The exposure frequency of the characters is a specific character. The data is stored in the database (9) and the analysis module 110 helps the module 11G to create a keyword and search for the keyword that is more detailed than the related keyword. The related keyword is combined with the category setting. Also, ^ the web-stomach file with the higher-weight phase keyword ^includes the relevant keyword with the : and unique H field_discovery new %. The squatting on the cheeks combines the documents on his website, and provides these documents to the aforementioned website. For example, in the document of the second page 200832268, the advertisement matching module 160 determines the category of the corresponding webpage file by using the characters included in the corresponding webpage file and category setting, and matches the advertisement of the advertiser, and the advertiser sets the judgment according to the corresponding webpage file. category. That is, the advertisement matching module 160 reads each weight of the character (related keyword) according to each category of the category settings stored in the database 15〇, and the characters are included in the network file, and Multiply the number of exposures for each relevant keyword based on a weight value. In addition, the category including the highest weight is the category of the corresponding webpage file, and the advertisement of the advertiser is combined with the related webpage file, and the advertiser issues an advertisement according to the corresponding category. As described above, one or more webpage files are extracted from the plurality of websites, each of which includes a primary keyword or a related keyword that conforms to a preset criterion according to each category, and the complex digital element is extracted from the webpage file, and is calculated. The weight of each character in the corresponding category. The weight value of the character conforms to a preset criterion, and a specific character is selected as a related keyword in the corresponding category to form a category setting accompanying the primary keyword. After the category is set, the process of forming the relevant keywords will be repeated periodically or irregularly to properly process the newly generated items of the corresponding category. The category setting formed according to the above description is an advertisement for conforming to the webpage file, so the weight of the character included in the webpage file is added to each other in each category to set the result of the category having the highest value as the corresponding webpage file. The category allows advertisements registered to advertisers of this category to be combined with webpage files to match the most appropriate advertisements and webpage files. According to the present invention, an advertisement most relevant to the context of the corresponding web page file can be displayed on a web page file of the web. Also, if a category is selected, new items related to the corresponding category can be appropriately processed. While the present invention has been described in terms of the preferred embodiments of the present invention, as illustrated in the foregoing description of the claimed invention, Modification, addition and replacement. BRIEF DESCRIPTION OF THE DRAWINGS [0009] The above and other objects, features, and advantages of the present invention will become more apparent from the aspects of the invention. An illustration of an example of a keyword ad.

Figure 2 is a schematic diagram conceptually showing an advertising method based on one of the present inventions. Only e is listed as a category. Fig. 3 is a schematic diagram of an embodiment of the present invention in accordance with the present invention. Based on the broad view, FIG. 4 is a schematic diagram conceptually showing a category-based advertising method in accordance with an embodiment of the present invention, showing that two or more categories match a single character. Figure 5 shows a schematic diagram of a method of calculating the weight of a single character, which is included in two or more categories. Figure 6 is a diagram showing an illustration of the different weights to a word based on the position of the characters in the file. [Main component symbol description] 61 · Title portion 62, 64 · Body portion 65: Background material portion 100: Advertising system 110: Analysis module 120 Keyword extraction module 130: Weight calculation module 19 200832268 131 : Exposure Concentration calculation module 132: station weight calculation module 133: trust level calculation module 134: position weight calculation module 135 · first weight calculation module 136: second weight calculation module 140: category matching module 141 : related keyword determination module 142 : category setting generation module 150 : database 160 · · advertisement matching module

20

Claims (1)

200832268 X. Patent application scope: 182: Declaring the advertising system as the basis, the system includes: a group for extracting one or more webpage files, each webpage file including = reading (four) files representing each category The primary key; the key sub-extraction kernel group' is used to extract one or more keywords included in the extracted file of the 4b network; a 〗 〖Bei, ΐ3:: 杈 is used based on the character Exposure feature relative to correspondence The weight of each character, the webpage files include the primary keywords of the pair of library categories; a category matching module for setting one or more characters to match the preset criteria in each corresponding category a plurality of related keywords of the corresponding category, and generating one or more category settings according to the keywords and their weights; and a message matching module for using the file included in the webpage and the category setting The word in the file is the same as the page file, and an advertisement registered as an advertiser of the category matches the corresponding web page file. 2. The system of claim 2, wherein the exposure characteristics comprise: an average exposure solution and an "exposure needle subtraction", wherein the fresh average exposure frequency represents a frequency at which a character is exposed in each file, the exposure concentration The degree is calculated using the ratio of the number of documents having the characters of the exposure to the total number of documents. The system described in claim 2, wherein the calculation of the weight of the exposure concentration is calculated. The module assigns a higher weight to a character included in a web page file, the web page file being extracted from a site, and the station has a higher weight relative to each category. 21 200832268 The system of light sets, wherein when the position of the exposure is calculated, the phase is difficult to be placed in the character, and the system described in item 4, wherein the position of the character - m: ^ part , the body part and the background material part of the position of the lion part, the financial part can be background (4) part and reduce ^ ^ as claimed in the scope of the patent scope 帛 1 item, wherein the category matching element matches the corresponding The plural-related two-parameters of the category are in the corresponding category, and the weight is equal to or greater than the system described in item 1 of the pre-set patent patent range, wherein the category matches the yuan real two = 1: fixed character, compared to In this category, the specific word is more than 1 ship, the age of the ship is regular, and the net is as described in item 1 of the patent scope, where the new one is added, and the same station is kept, the system updates the Category setting. 10. If the system described in item (4) i of the patent application is filed, the system will update the remuneration settings when the document is filed. w ' 11. According to the system described in item J of the patent application, one page file is extracted, and the network element includes plural keywords, which are included in the category settings and are consistent with the preset . The 'sub' is 12. The relevant keyword of the system sighing as described in the patent scope fi item includes the #1 level - two two: two precautions are higher than the corresponding _ setting - the default level. And the temple in the 22 or 32,32,268, the t/ltt patent paradigm (4), the pure one, which meets a preset: the test (4) keyword includes only one of the categories that exist only in one category. As a basic advertising method, one or more webpages are extracted, and a method is used to include a +#+Bei file, and the parent-webpage file includes a primary key from a plurality of representative documents of the plurality of webpage files; - or more characters in the webpage files; The exposure feature of p t (four) t70 calculates the weight of the extracted parent relative to the corresponding category, and the webpage files include the primary key of the corresponding category; and the child or the child element is such that one of the preset criteria in each corresponding category meets Corresponding to the plural _ _ word of _, and generating one or more category settings according to the age keyword and its weight; and determining the category of each web page file using the characters included in the web page file and the category setting, An advertisement registered as an advertiser of the category is matched with the corresponding webpage file. 15. The method of claim 14, wherein the exposure characteristics comprise an average exposure frequency and a degree of exposure concentration, the average exposure frequency representing a frequency at which a character is exposed in each file, the exposure The degree of concentration is calculated using the ratio of the number of files having the characters of the exposure light to the total number of files. The method of claim 15, wherein when calculating the exposure concentration level, the weight calculation module specifies a higher weight to include a character in a webpage file, and the webpage file is extracted from One station, the station has a higher weight than each category. 23. 200832268 a method of the 15th patent scope, Α中^10-module module designation-different weight: upper position 17 妓 妓, shot-character position of the two ΐ: fine, the The body portion and the background material portion reduce the matching mode Lr by the method described in item 14 wherein the category is $'4 a tn eight: whether the matching of the corresponding category of the plural correlation key is preset - weight, in In a corresponding category, the weight is equal to or greater than 乂〇: as in the method of claim 14 of the patent application, and further includes the step of setting a $7L, wherein the category matching module specifies - I high weight is the special scorpion Second, compared to __, the specific word it exists only in the - category. Do this = please refer to the method described in item 14 of the patent scope, and include the steps to periodically update these category settings. 22. The method of claim 3, further comprising the step of updating the category settings by a new ',, station, ¥' system. The method H of the supplementary '14' includes the step of updating the category settings when the new page is added. 24. The method of claim 14, further comprising (d) wherein the network 1 file includes a plurality of related key sub-keys included in the category settings and conforming to a predetermined criterion. 25. The method of claim 14, wherein the phase _ key includes a weight of -level, and the i is equal to or equal to the face. - Preset level. The method of claim 14, wherein the related keywords that meet a predetermined reference include only one of the specific characters that are substantially present in one category. 25