JP5762480B2 - Information processing apparatus and method - Google Patents

Description

  The present invention relates to a so-called online advertisement.

  In recent years, various types of online advertisements have been known. For example, content-linked advertisements (for example, Patent Document 1) are used for advertising distribution destinations that match a keyword or interest category specified by an advertiser. It is displayed on a page (also called “distribution surface”). In addition to the web page, the page serving as a distribution surface is a page of content displayed by an application program (also referred to as “application”) of an information terminal such as a smartphone or a tablet PC.

  In this type of advertisement, if the relation between the ad text (called title & description etc.) displayed on the screen as an advertisement and the keyword specified by the advertiser is insufficient, the advertisement is less relevant to the content of the web page Is displayed and the click rate decreases. In order to prevent this problem, it has been devised to narrow down advertisements to be displayed based on the relationship between the web page of the advertisement distribution destination and the transition page when the advertisement text or advertisement is clicked.

Special table 2006-500698

  However, the ad text and the transition destination page are also a set of words based on the advertiser's designation in a broad sense and can be determined in any way as the advertiser intends. Sexual security was insufficient. For example, depending on how you use words in your ad text, recently trending words etc. are interspersed, so it will match and display many distribution aspects. Different advertisements are also possible.

  If many such ads are displayed, in addition to the problem of a decrease in the click rate, the ad viewers will be redirected to a different web page than expected, so the ads will be avoided There are also problems such as trends.

  An object of the present invention is to control the display amount of an advertisement based on an objective relationship with an advertiser.

  Based on the above object, an information processing apparatus according to one aspect (1) of the present invention relates to an advertisement storage unit that stores advertisement information including a designated word that is a word based on an advertiser's designation, and an advertiser. For each advertisement to be displayed, name storage means for storing a name, reference extraction means for extracting, as a reference word, a word that is included in a search result based on the stored name above a predetermined reference , Conformity determination means for determining the adaptability between the reference word extracted for the advertiser and the specified word included in the advertisement information, and control the display amount of the advertisement according to the determined adaptability Display control means.

  An information processing method according to another aspect (7) of the present invention is an advertisement that stores the advertisement information including a designated word that is a word based on an advertiser's designation for the advertisement, in which the above aspect is captured by a category of the method. A computer comprising storage means and name storage means for storing names relating to advertisers extracts, as reference words, words that are included in search results based on the stored names above a predetermined reference. A reference extraction process, and a determination process for determining a matching degree between the reference word extracted for the advertiser and the specified word included in the advertisement information for each advertisement to be displayed And a display control process for controlling a display amount of the advertisement according to the fitness.

  According to another aspect (2) of the present invention, in any one of the above aspects, the conformity determination means determines the conformity based on a closeness of a semantic distance between the reference word and the designated word. It is characterized by.

  According to another aspect (3) of the present invention, in any one of the above aspects, the advertisement information includes an advertisement text designated by an advertiser, and a plurality of advertisement candidates are extracted from the advertisement information together with a display probability for each candidate. For each extracted candidate for the advertisement, the matching determination unit determines the degree of matching with the reference word corresponding to the advertiser with the word in the corresponding advertisement text as the designated word. The display control means determines an advertisement to be displayed based on the display probability for each candidate and the determined fitness.

  According to another aspect (4) of the present invention, in any one of the above aspects, the display control unit determines, for each candidate, the number of designated words that match the reference word among the designated words in the advertisement text. Calculation means for calculating a calculated value by dividing by the number of the specified words in the advertisement text and multiplying the display probability, and determination means for determining an advertisement to be displayed from among the candidates based on the calculated value And.

  According to another aspect (5) of the present invention, in any one of the above aspects, a category storage unit that stores a business category in association with each advertiser is provided, and the conformity determination unit includes an advertisement in the category storage unit. For the advertisement in which the business category corresponding mainly is stored, the degree of conformity between the business category and the designated word included in the advertisement information is determined.

  According to another aspect (6) of the present invention, in any one of the above aspects, there is provided corresponding word storage means for storing words corresponding to each of the business categories, wherein the conformity determination means includes the business category and the designation. The degree of matching with a word is determined based on the amount of matching between the word stored in the corresponding word storage means corresponding to the business category and the word in the advertisement text designated by the advertiser. Features.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to control the display amount of an advertisement on the basis of objective relevance with an advertiser.

The functional block diagram which shows a structure about embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The flowchart which shows the process sequence in embodiment of this invention. The conceptual diagram which shows the example of the process in embodiment of this invention. The conceptual diagram which shows the example of the process in embodiment of this invention.

  Next, a mode for carrying out the present invention (referred to as “embodiment”) will be illustrated along the drawings. In addition, the assumptions common to the contents already described in the background art and problems are omitted as appropriate.

[1. Constitution〕
FIG. 1 is a configuration diagram of this embodiment. The server 1 is a means for providing an advertisement to be displayed on a web page or other page to the terminal T via the communication network N (Internet, mobile phone network, etc.), and the page itself may be provided from another server. .

  The server 1 includes, as a computer configuration, an arithmetic control unit 6 such as a CPU, a storage device 7 such as a main memory and an auxiliary storage device, and a communication device 8 (such as a communication device and a communication adapter) with the communication network N. Have. The terminal T also has a computer configuration (not shown), although the specifications are different. A large number of terminals T actually exist according to the number of users, and are portable information terminals such as smartphones and tablet PCs, and other personal computers.

  In the server 1, each element shown in FIG. 1 is realized by the calculation control unit 6 executing a computer program (not shown) stored in the storage device 7.

  Among the realized elements, the information storage means is not limited to the so-called local storage in the server 1 but may be remote storage by network computing (cloud) or the like. In addition, the storage means shown in the present application is a main unit shown for convenience of explanation, and actually includes functions such as input / output and management that accompany information storage, and the unit of configuration is divided or They may be integrated or other storage means such as a work area may be used as appropriate.

  Note that the arrows in the figure (for example, FIG. 1) supplementarily indicate the main direction of the flow of data, control, etc., and do not mean the denial of other flows or the limitation of the direction. In addition, each means other than the storage means is a processing means for realizing and executing the information processing functions and operations described below, but these are functional units for convenience of explanation, and actual hardware elements and software. It doesn't matter if it matches the module.

  Among the storage means, the advertisement storage means 15 is means for storing advertisement information including a designated word that is a word based on the designation of the advertiser for the advertisement submitted by the advertiser (for example, FIG. 2). The information includes an advertisement text (including a designated word) designated by the advertiser. Further, the name storage means 25 is a means for storing a name related to an advertiser (an organization name such as a company name, a business name, etc.) (for example, FIG. 3).

  The category storage unit 55 is a unit that stores an objective business category in association with each advertiser, and FIG. 3 shows the name storage unit 25 and the category storage unit 55 together. The business category is not based on the advertiser's own self-reporting, but is given by an advertising company such as the operator of the server 1 based on a stock information magazine or other reliable information source. Corresponding word storage means 66 is means for storing words corresponding to each business category (for example, FIG. 4).

[2. Action)
FIG. 5 is a flowchart illustrating a processing procedure in which the server 1 configured as described above determines a predetermined number of advertisements to be displayed on the distribution surface. FIGS. 6 and 7 are conceptual diagrams illustrating the process from extraction of candidates to calculation of an adjusted probability (described later), which is a calculated value used to determine the advertisement.

[2-1. Extract candidates
In order to determine an advertisement to be displayed on the distribution surface serving as a display destination, first, the candidate extraction unit 50 matches the word in the advertisement text corresponding to the designated word and the distribution surface from the advertisement information stored in the advertisement storage unit 15. A plurality of advertisement candidates are extracted together with the display probabilities for each candidate by (compatibility determination) or the like (step S11).

  As a simple example, two candidates shown in FIG. 2 are extracted as one advertisement candidate to be displayed on the distribution surface relating to travel, and the display probability calculated based on the degree of matching with the distribution surface, the bid price, etc. at the time of extraction Are the same 50%.

  Then, for each extracted advertisement candidate (steps S12 to S18), the fitness is determined as follows. Steps S14 and S17 will be described later, and here, a case where there is no information on the business category (step S14 and the rightmost column in FIG. 3) will be described.

[2-2. Extraction of specified words)
Here, first, the conformity determination means 30 extracts words constituting the advertising text as designated words from the advertising text of each advertisement by morphological analysis or the like (step S13). At this time, in the advertisement text, the name of the advertiser, such as the company name or service name, may be excluded from the extraction target.

  For example, for the advertisement with the advertisement ID “A15-02” shown in FIG. 2 (referred to as the “first advertisement”) (FIG. 6), the company name from the advertisement text “Cheap domestic travel! As a result of morphological analysis and syntactic analysis of the part “excluded at low prices for domestic travel!”, The included words are “low price”, “domestic”, and “travel”.

  On the other hand, for the advertisement of another advertisement ID “B38-72” shown in FIG. 2 (referred to as “second advertisement”) (FIG. 7), the advertisement text “Cheap domestic travel! Analyzing “Let's apply Pad” in the same way, it is assumed that the words included are “Cheap”, “Domestic”, “Travel”, “Credit Card”, “O Pad” and “Let's Make”.

[2-3. (Extraction of reference words)
Further, the reference extraction unit 20 accesses the search server 2 based on the name stored in the name storage unit 25 for the advertiser of the advertisement, and performs a predetermined search such as a web search by a search engine using the name as a search key. A search is performed, and words that are included in the search result above a predetermined reference are extracted as reference words (step S15).

  For example, for the first advertisement (FIG. 6), the name (FIG. 3) relating to the advertiser (advertiser ID “A15”) is “xx travel agency”. It is assumed that a word that appears more than a predetermined in a web search result using the name “XX travel agency” as a search term is “cheap, domestic, overseas, travel, reservation, etc.”. These words become reference words.

  On the other hand, for the second advertisement (FIG. 7), a word (reference word) that appears more than a predetermined number in the web search result with the name “△△ Finance” relating to the advertiser (advertiser ID “B38”) as a search term is “Credit card, loan, screening, interest rate…” etc.

  The name related to the advertiser that is the basis of the reference word is not limited to the organization name such as the official company name of the advertiser (for example, “XX travel company”), but the name of the service (for example, “XX travel”), A common name may be used. Examples of words that have appeared more than a predetermined number are words with the total number of appearances in the top 100 web pages whose URLs are included in the search results by name, such as the top 10 and the 20th. The number of words that appear more than a predetermined number may be varied according to the number of search hits by the name.

[2-4. Determination of fitness and determination of advertisement)
Subsequently, the matching determination means 30 matches the degree of matching between the reference word extracted as described above for the advertiser and the designated word included in the advertisement information for each advertisement to be displayed, that is, the extracted advertisement candidate. Is determined (step S16). Here, the designated words included in the advertisement information are words included in the advertisement text as described above.

  That is, for each extracted advertisement candidate, the matching determination unit 30 determines the matching degree with the reference word corresponding to the advertiser, using the word in the corresponding advertisement text as the designated word. For example, when 10 designated words are extracted from an advertisement sentence having a maximum length of 33 characters by morphological analysis, and the reference word based on the name related to the advertiser is 20 words, the match between the 10 designated words and the 20 reference words Sex (eg, all or part of the match) can be determined by 200 comparison calculations.

  Further, the display control means 40 determines one or two or more predetermined number of advertisements to be displayed based on the display probability for each candidate and the determined adaptability (step S19), and the determined adaptability. Control the display amount of advertisements according to the degree.

  More specifically, the display control unit 40 includes a calculation unit 41 and a determination unit 42, and the calculation unit 41 calculates the number of designated words that match the reference word among the designated words in the advertisement sentence for each candidate. The calculated value (referred to as “adjusted probability”) is calculated by dividing by the number of designated words and multiplied by the display probability, and the determining means 42 determines the candidate contents based on the calculated adjusted probability. Decide which advertisements to display.

  For the first advertisement (FIG. 6), the three designated words are all advertiser reference words (indicated by ellipses in FIG. 6). The number of 3 words is divided by the number of 3 words specified in the ad text to obtain a coefficient of 1.0 (100%), which is adjusted by multiplying this by, for example, the display probability (for example, 50%) The probability remains 50%. In the advertisement text, the name of the advertiser, such as the company name or service name, may be ignored in the determination of compatibility with the reference word, or may be counted as the number of matching words.

  As for the second advertisement (Fig. 7), the designated words “Cheap”, “Domestic”, “Travel”, “Credit Card”, “○” Among the six words “Pad” and “Let's guess”, the advertiser's reference word is only one word (indicated by an ellipse in FIG. 7), and the other five words are not reference words (indicated by x in FIG. 7).

  Therefore, even if the display probability of the second advertisement is 50%, the adjusted probability is about 8.3% when multiplied by 1/6 as a coefficient, which is inferior to the adjusted probability of 50% of the first advertisement. To do. For example, when a sufficiently large number of times are displayed at random according to the ratio of the adjusted probability, the display frequency ratio between the first advertisement and the second advertisement is expected to be approximately 6: 1.

  In the above example, the fitness is determined using the number and ratio of the designated words that match the reference word. However, the fitness may be determined based on the closeness of the meanings of the reference word and the designated word. In this case, the suitability determining means 30 calculates the closeness of the semantic distance between the reference word and the designated word using a so-called semantic dictionary or the like, and determines the suitability based on that. A conventionally known technique may be used for the semantic distance and the semantic dictionary.

  For example, in the case of 10 specified words and 20 reference words, the closest reference word can be specified for each specified word by calculating the semantic distance 200 times in total for all combinations. The total score of the closeness of the semantic distance to the nearest reference word for every specified word may be used as the fitness.

[2-5. Use of business categories)
Instead of extracting the reference word each time, it is possible to determine the degree of matching between the business category previously associated with the advertiser and the designated word. Specifically, the conformity determination unit 30 determines whether the business category corresponding to the advertiser is stored in the category storage unit 55 (the rightmost column in FIG. 3) (step S14: “YES”). The degree of matching with the designated word included in the advertisement information is determined (step S17).

  The matching degree between the designated word of the advertisement and the business category is, for example, a word representing the business category itself (for example, “travel” for the travel industry, “finance”, “loan”, “loan”, etc.) for the finance industry, However, in this embodiment, the determination is made by using one or more words (referred to as “corresponding words”) associated with the business category in advance.

  Specifically, the conformity determination means 30 determines the degree of conformity between the business category and the designated word, the corresponding word (for example, FIG. 4) stored in the corresponding word storage means 66 corresponding to the business category, and the advertisement. A determination is made based on the amount of matching with the word in the advertisement text designated by the main (step S17). The corresponding word may be manually set by a human, but the reference word (step S15) extracted from the name search result may be stored and used as a corresponding word.

  For example, the business category corresponding to an advertiser “XX travel agency” is “travel business” (FIG. 3), and the word corresponding to this business category is “cheap, domestic, overseas, travel, reservation ...”, etc. In this case (FIG. 4), it is not necessary to extract a reference word by web search based on the advertiser name, and the process of determining an advertisement can be speeded up.

[3. effect〕
As described above, in the present embodiment, by using many words in the search result of the name related to the advertiser as an objective criterion, the display amount of the advertisement according to the compatibility with the designated word based on the designation of the advertiser ( For example, the display amount of the advertisement can be controlled based on the objective relevance with the advertiser as shown in FIGS.

  Further, in the present embodiment, it is possible to make an appropriate determination even if there is a blur in the written expression of the word, by determining the degree of matching based on the semantic distance between words, not on the coincidence of words.

  Further, in the present embodiment, by determining the advertisement to be displayed by adding the compatibility between the word in the advertisement text and the reference word to the display probability for each extracted advertisement candidate (FIGS. 6 and 7), Even if there is an advertisement including a distracted word in the advertisement text, it is possible to suppress a situation where the advertisement is displayed on a variety of distribution surfaces that match any word, and the quality of the displayed advertisement is improved.

  Further, in the present embodiment, an advertisement to be displayed with an adjusted probability obtained by multiplying the display probability of each extracted candidate by the ratio of words that match the reference word among the designated words in the advertisement text is determined (FIG. 6, FIG. 7). Accordingly, the function of the present invention can be added as a post-processing to a conventional advertisement selection system that extracts advertisement candidates to be displayed with a display probability, and the application target can be expanded.

  Further, in the present embodiment, an advertisement having a business category previously associated with an advertiser based on a search result or manually is used for determining the fitness (steps of FIGS. 3 and 5). S14 and S17), web search and reference word extraction each time can be omitted, and the processing can be made efficient.

  Further, in the present embodiment, the business category (FIG. 3) and the business category in the semantic dictionary or the like are determined by determining the degree of matching based on the matching amount between the word group (FIG. 4) associated in advance and the word in the advertisement text. The complicated process of determining the semantic distance to the specified word can be omitted.

[4. Other embodiments]
In addition, the content of the said embodiment and figure is only an illustration, and the presence or absence and arrangement | positioning of each element, the order and content of a process, etc. can be changed suitably. For this reason, this invention also includes the modification illustrated below and other embodiment other than that. For example, advertisements to which the present invention is applied are not limited to content-linked advertisements, but may be search-linked advertisements, behavioral targeting advertisements, etc., and display advertisements such as banner advertisements using images and videos, not limited to text advertisements. But you can.

  In addition, the specified word is not limited to the word in the ad text, the word specified by the advertiser to narrow down the distribution side of the distribution destination as the condition for displaying the advertisement, the word included in the transition destination web page when clicking the advertisement, etc. But you can. The degree of conformity is not limited to the determination based on the closeness of the semantic distance between the reference word and the designated word, but may be determined based on a match between all or a part of the reference word and the designated word, or a match based on the synonym dictionary. .

  Moreover, each aspect of the present invention can be understood as other categories (methods, programs, systems including terminals, etc.) that are not specified. In the category of method or program, “means” shown in the category of apparatus is appropriately read as “process” or “step”. In addition, all or any part of “means” can be read as “part” (unit, section, module, etc.).

  Also, the processes and steps shown in the embodiment can be changed by changing the order, executing some of them collectively, or executing them part by part. In addition, hardware elements that implement and execute individual means, processes, and steps may be common, or may differ for each means, process, step, or timing.

  The individual means shown in the present application may be realized by calling a function provided by an external server by an API (Application Program Interface) or network computing (so-called cloud or the like). Furthermore, each element such as means relating to the present invention may be realized by other information processing mechanisms such as a physical electronic circuit as well as a computer control unit.

1 Server 2 Search Server 6 Arithmetic Control Unit 7 Storage Device 8 Communication Device 15 Advertising Storage Unit 20 Reference Extraction Unit 25 Name Storage Unit 30 Suitability Determination Unit 40 Display Control Unit 41 Calculation Unit 42 Determination Unit 50 Candidate Extraction Unit 55 Category Storage Unit 66 Corresponding word storage means N Communication network T Terminal

Claims (7)

An advertisement storage means for storing advertisement information including a designated word that is a word based on an advertiser's designation for the advertisement;
Name storage means for storing names related to advertisers;
Reference extraction means for extracting, as a reference word, a word included in a search result based on the stored name above a predetermined reference;
For each of the advertisements to be displayed, suitability determination means for determining the suitability of the reference word extracted for the advertiser and the specified word included in the advertisement information;
Display control means for controlling the display amount of the advertisement in accordance with the determined fitness,
An information processing apparatus comprising:   The information processing apparatus according to claim 1, wherein the matching determination unit determines the matching level based on a closeness of a semantic distance between the reference word and the designated word. The advertising information includes an advertising text designated by the advertiser,
Candidate extraction means for extracting a plurality of advertisement candidates from the advertisement information together with a display probability for each candidate;
The suitability determining means determines, for each extracted advertisement candidate, a matching degree with the reference word corresponding to the advertiser, with the word in the corresponding advertisement text as the designated word,
The information processing apparatus according to claim 1, wherein the display control unit determines an advertisement to be displayed based on the display probability for each candidate and the determined fitness. The display control means includes
For each of the candidates, a calculation for calculating a calculated value by dividing the number of designated words that match the reference word among the designated words in the advertisement text by the number of the designated words in the advertisement text and multiplying the display probability. Means,
Determining means for determining an advertisement to be displayed from among the candidates based on the calculated value;
The information processing apparatus according to claim 3, further comprising: Category storage means for storing business categories in association with each advertiser,
The conformity determining means determines the degree of conformity between the business category corresponding to the advertiser stored in the category storage means and the business category and the specified word included in the advertisement information. The information processing apparatus according to claim 1, wherein the information processing apparatus is an information processing apparatus. Corresponding word storage means for storing words corresponding to each business category,
The conformity determination means includes a word stored in the corresponding word storage means corresponding to the business category and a word in an advertisement sentence designated by the advertiser for the degree of conformity between the business category and the designated word. The information processing apparatus according to claim 5, wherein the determination is based on the amount of coincidence with the information processing apparatus. An advertisement storage means for storing advertisement information including a designated word that is a word based on an advertiser's designation for the advertisement;
Name storage means for storing names related to advertisers;
A computer with
A reference extraction process for extracting, as a reference word, a word included in a search result based on the stored name above a predetermined reference;
A matching determination process for determining a matching degree between the reference word extracted for the advertiser and the designated word included in the advertisement information for each advertisement to be displayed;
Display control processing for controlling the display amount of the advertisement according to the determined fitness,
The information processing method characterized by performing.

Images