JP7170959B2 - DATA GENERATION DEVICE, GENERATION METHOD AND PROGRAM THEREFROM FOR EXTRACTING TVCM - Google Patents

Description

The present invention relates to a data generation device, a generation method, and a program therefor for making it possible to extract TVCM.

Regarding commercials aired on TV (hereinafter referred to as "TV commercials"), there are situations where data on when and what kind of commercials were aired is useful. is one. This is because past TV commercials can serve as a reference when producing TV commercials, and can also serve as a reference for determining the broadcast date and time. This is the same for companies that provide services as an advertising agency to companies that are considering using TVCM as customers.

However, although it is easy to refer to current or past TV commercials based on the talents used, the date and time of airing, etc. Difficulties arise when we stand in the perspective of whether In other words, there is no method for efficiently identifying such TV commercials from current or past broadcast data. If the name of the competitor company is clear, it is easy to identify the TV commercial by the company. They are rare.

The present invention has been made in view of such problems, and its purpose is to extract TV commercials by a first company or a second company related to its business from current or past TV commercials. The present invention relates to a data generation device, a generation method, and a program therefor.

In order to achieve such an object, a first aspect of the present invention selects a TVCM by a first company or a second company related to its business from one or more current or past TVCMs. A method for generating data to be extractable, comprising: a step of a computer reading each broadcast data of one or more of the current or past TV commercials; a step of inputting at least one of a company name and a product name, a service name, or a brand name included in the broadcast data into a search engine as a keyword; The website displayed first and the domain that can be owned by a corporation that is one level below the general top-level domain or the domain that can be owned by a corporation that is two or one level below the country code top-level domain a step of acquiring one or more character strings described in any meta tag or title tag of the same website; and storing.

A second aspect of the present invention is characterized in that, in the first aspect, the metatag includes a description metatag or a keywords metatag.

A third aspect of the present invention is characterized in that, in the first aspect, the metatag includes a description tag of OGP or twitter.

Further, according to a fourth aspect of the present invention, in the first aspect, the one or more character strings are at least one of one or more nouns extracted by analyzing the content attribute description of the metatag or title tag. It is characterized by being a part.

Further, according to a fifth aspect of the present invention, in any one of the first to fourth aspects, at least part of the one or more character strings does not include the keyword.

Further, according to a sixth aspect of the present invention, in any one of the first to fifth aspects, the input is performed using a company name, product name, service name, or brand name included in the broadcast data as a keyword. Characterized by

Further, according to a seventh aspect of the present invention, in any one of the first to sixth aspects, the input is performed using a service name or brand name included in the broadcast data as a keyword, and the service name or brand name is , the service name or brand name of an Internet service.

Further, according to an eighth aspect of the present invention, in any one of the first to seventh aspects, the acquisition is displayed first by excluding a predetermined one or more websites from the search results. and the domain that can be owned by a corporation that is one level below the general top-level domain, or the website that has the same domain that can be owned by a corporation that is two levels or one level below the country code top-level domain It is characterized by performing against.

A ninth aspect of the present invention is characterized in that, in the eighth aspect, the predetermined one or more websites include at least one of an EC site, a recruiting site, an encyclopedia site, or a dictionary site. .

In addition, a tenth aspect of the present invention provides a data structure for extracting a TVCM by a first company or a second company related to its business from one or more current or past TVCMs. and is a data structure generated by the generation method according to any one of the first to ninth aspects.

In addition, in the eleventh aspect, the computer stores data for extracting TVCM by a first company or a second company related to its business from one or more current or past TVCM. A program for executing a generation method, the generation method comprising the steps of: reading each broadcast data of one or more of the current or past TVCMs by the computer; a step of inputting at least one of a company name and a product name, a service name, or a brand name included in the broadcast data into a search engine as a keyword for each identifier; The website displayed first except for the results of the above and the domain that can be owned by a corporation one level below the general top-level domain, or owned by a corporation two or one level below the country code top-level domain a step of acquiring one or more character strings described in any meta tag or title tag of a website having the same possible domain; and storing in association with the data.

In addition, in the twelfth aspect, a data generation device for extracting TVCM by a first company or a second company related to its business from one or more current or past TVCM reading each broadcast data of one or more of the current or past TVCM, and at least one of a company name and a product name, a service name, or a brand name included in the broadcast data for each CM identifier included in the broadcast data; is entered as a keyword into a search engine, and the website displayed first by excluding the results displayed as advertisements in the search engine from the search results and the corporation one level below the general top level domain can be owned domain, or a domain that is two levels below or one level below the country code top-level domain and that can be owned by a corporation has the same website. , at least part of the one or more character strings are stored in association with the broadcast data.

According to one aspect of the present invention, at least one of a company name, a product name, a service name, or a brand name included in the broadcast data is used as a keyword for each CM identifier in the broadcast data of one or a plurality of current or past TV commercials. By using the obtained search engine search results to obtain a character string that links the company that broadcast the broadcast data or a company that is related to that business or its business, and stores it in association with the broadcast data , data from which TV commercials by a first company or a second company with relevance to its business can be extracted.

1 is a diagram showing a generating device according to an embodiment of the invention; FIG. It is a figure which shows the flow of the production|generation method concerning one Embodiment of this invention. It is a figure which shows a part of source code of the website concerning one Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described in detail.

FIG. 1 shows a generating device according to one embodiment of the invention. The generation device 100 can read out each broadcast data from a database 110 in which broadcast data of one or a plurality of current or past TVCM are stored, and a server 120 that provides a search engine on an IP network such as the Internet. can communicate.

Each piece of broadcast data stored in the database 110 includes a CM identifier that identifies the CM, the name of the company broadcasting the CM, and the name of the product, service, or brand (hereinafter referred to as "products, etc.") that is the target of the CM. including. A CM may have a plurality of different versions even if it targets the same product or the like, and in such cases, different CM identifiers may be assigned. Only one of the company name and product name may be included, and abbreviations of these formal names may be used. In addition, the broadcast data may further include broadcast start date and time, broadcast end date and time, number of seconds of CM, names of entertainers who appeared, names of songs used, artist names of songs used, and the like.

The device 100 includes a communication unit 101 such as a communication interface, a processing unit 102 such as a processor and a CPU, and a storage unit 103 including a storage device or storage medium such as a memory and a hard disk, and stores programs for performing each process. It can be constructed by executing and may include one or more devices or servers. Also, the program may include one or more programs, and may be recorded on a computer-readable storage medium to be a non-transitory program product. The program can be stored in a storage device or storage medium accessible from the storage unit 103 or the device 100 and executed by the processing unit 102 .

In the following description, it is assumed that the device 100 reads broadcast data from the database 110, but the broadcast data stored in the database 110 is stored in the storage unit 103 of the device 100 or in a storage device or storage medium accessible from the device 100. You can keep it.

First, the device 100 reads out each broadcast data of one or a plurality of current or past TVCM (S201). Next, the device 100 inputs at least one of the company name, product name, service name, and brand name included in the broadcast data into the search engine as a keyword for each CM identifier included in the broadcast data (S202).

A keyword can be input to the search engine, for example, by the device 100 operating a web browser and inputting the keyword in the input field of the search engine. The device 100 can operate a web browser stored in the storage unit 103 or in a storage device or storage medium accessible from the device 100, and code written using a library for operating the web browser as necessary. The web browser can be made operable by executing Examples of web browsers include Chrome (registered trademark). The input of keywords to the search engine may be operated headless with a web browser. The input of keywords to the search engine and the reception of search results can also be performed by transmitting and receiving data according to the HTTP or HTTPS protocol instead of operating the web browser. An example of a library that can be used to transmit and receive such data is curl (trademark).

Next, from the search results of the search engine, the device 100 excludes the results displayed as advertisements in the search engine, and acquires one or more character strings described in the metatags of the first displayed website. (S203). It is also conceivable to acquire the character string described in the title tag instead of the meta tag. Beautiful Soup (trademark) is an example of a library for obtaining descriptions of metatags or title tags from website html files.

Examples of metatags can include the description metatag or the keywords metatag. These meta tags describe words or expressions that the publisher of the website considers to be relevant to their company or their business, and are used by search engines to identify the websites of other companies that are relevant to their company or their business. Since it is intended to be compared with the website or to be displayed higher than the website of the other company, companies or products that share the same character string described here The inventors have found that they can be evaluated as competitors and at least as companies or businesses related to them or their businesses. Other examples of meta tags may include an OGP description tag, a twitter (registered trademark) description tag, and the like. Further details of string acquisition will be described later with an example.

Then, the device 100 stores at least part of the one or more character strings obtained in association with the broadcast data of the CM identifier (S204). Storage may be in database 110 or may be in a storage device or medium accessible from storage 103 or device 100 . Also, at least part of the one or more character strings to be stored may not include the input keyword. The company names or product names used as keywords represent individual companies or products, and by removing such words, the character string more expresses the relationship between companies or products. can be extracted.

In this way, by associating at least a part of the one or more character strings with each CM identifier to broadcast data of one or more TV commercials currently or in the past, the first company can be identified from the broadcast data. Alternatively, data is generated that can extract TVCM by a corporation or a second corporation or corporation that has relevance to its business.

Some broadcast data of TVCM are available for purchase. Some broadcast data includes a classification code representing the content of the commercial. For example, the content of a commercial may be identified by large, medium, or small categories such as "food", "seasoning", and "sauce". However, for example, various Internet services are currently being created, and TV commercials by companies that provide such services are increasing. In fact, at present, classifications such as "personal computer/AV", "others", and "website" are collectively given to various Internet services, and they cannot function as classifications for distinguishing related products. The present invention is particularly useful for products that change rapidly and that are difficult to classify appropriately, such as Internet services, which are services provided on the Internet.

Consider extracting current or past TVCM using data generated by the generation method according to the present embodiment. Using the generated data as a database, when searching the database with the character string "sharing" using the example of "Hacobell" shown in FIG. are acquired, and at least part of the company name, product name, broadcast date and time, talent name, etc. included in the broadcast data can be displayed on the user's display screen.

In the above description, the device 100 targets the first website displayed by excluding the results displayed as advertisements in the search engine from the search results of the search engine. The website displayed on the website and a domain that can be owned by a company or corporation that is one level below a general top-level domain (gTLD) such as "com", or a two-level country code top-level domain (ccTLD) such as "jp" Websites with the same domain that can be owned by companies or corporations one level below or one level below may be targeted. In the case of ccTLDs, companies or corporations may be able to own domains one level lower such as "abc.jp" in addition to domains two levels lower such as "abc.co.jp". When a company name or product name is entered as a keyword, scraping the meta tag or title tag of the official website of the product is considered to give high accuracy, but it is not always displayed first. It does not deviate from the spirit of the present invention to target websites of the same domain.

In addition, it is preferable to determine target websites by excluding EC sites such as "amazon.com (registered trademark)" from the search results. This is because such EC sites may be displayed at the top of search results. Similarly, encyclopedia sites or dictionary sites such as "wikipedia (registered trademark).org" are preferably excluded. In addition, it is preferable to exclude job search sites as they may be displayed at the top of the search results.

In addition, unless there is a description of "only" such as "based on XX only", "only in response to XX", or "in the case of XX only", in this specification, additional information Note that it is assumed that . Also, as an example, it should be noted that the statement "when a, do b" does not necessarily mean "when a, do b", unless explicitly stated otherwise.

Also, just in case, any method, program, terminal, device, server, or system (hereinafter "method, etc."), even if there is an aspect that operates differently from the operations described in this specification, each of the present invention Aspects are directed to acts that are identical to any of the acts described herein, and that there are acts that are different from those described herein, the methods, etc. It should be added that it is not out of the scope of the embodiments.

Get string details
Figure 3 shows the first website displayed after excluding advertisements after searching for the service name of "Hacobel (registered trademark)", which is a service provided by the applicant, using the search engine "Google (registered trademark)". A part of the source code of the official website of the service is shown.

When a character string is obtained from the title tag, morphological analysis can be performed on the description of the content attribute of the title tag to extract one or more character strings corresponding to nouns. In the example of FIG. 3, character strings such as "distribution", "schelling", and "platform" can be acquired. This also applies to description meta tags, OGP description tags, and the like. The keywords meta tag may also be obtained by morphological analysis in the same way. You can do the analysis.

100 device 101 communication unit 102 processing unit 103 storage unit 110 database 120 search engine server

Claims (8)

A method for generating data for extracting TVCMs by other companies from one or more current or past TVCMs,
a step in which a computer reads each broadcast data of the current or past one or more TV commercials;
a step in which the computer inputs at least one of the company name and product name, service name, or brand name included in the read broadcast data as a keyword into a search engine;
A step in which the computer acquires one or more character strings described in any meta tag or title tag of a website displayed by excluding results displayed as advertisements in the search engine from search results;
and a step in which the computer stores at least part of the one or more character strings, at least part of which does not include the keyword, in a searchable manner in association with the broadcast data. Method. 2. The method of claim 1, wherein the metatags include description metatags or keywords metatags. 2. The method of claim 1, wherein the meta tags include OGP or twitter description tags. 4. The method according to any one of claims 1 to 3, wherein said input is performed using a company name, product name, service name, or brand name included in the read broadcast data as a keyword. The above acquisition can be held by the meta tag or title tag of the website displayed by excluding the results displayed as advertisements on the search engine from the search results, or by a corporation one level below the website and the general top-level domain One or more character strings described in any meta tag or title tag of a website with the same domain that can be owned by a corporation that is two levels below or one level below the country code top-level domain. 5. A method according to any one of claims 1 to 4, characterized in that Said acquisition is the domain or country code that can be owned by the corporation one level below the website displayed first by excluding one or more predetermined websites from the search results and the general top-level domain 6. The method of claim 5, wherein the website is either two levels below the top-level domain or one level below the top-level domain and the corporate-owned domain is the same. A program for causing a computer to execute a data generation method for extracting TVCMs by other companies from one or more current or past TVCMs, wherein the generation method comprises:
a step of the computer reading each broadcast data of the current or past one or more TV commercials;
a step in which the computer inputs at least one of the company name and product name, service name, or brand name included in the read broadcast data as a keyword into a search engine;
A step in which the computer acquires one or more character strings described in any meta tag or title tag of a website displayed by excluding results displayed as advertisements in the search engine from search results;
and a step in which the computer stores at least part of the one or more character strings, at least part of which does not include the keyword, in a searchable manner in association with the broadcast data. program. A data generation device for extracting TVCMs by other companies from one or more current or past TVCMs,
reading each broadcast data of the current or past one or more TVCM;
At least one of the company name and product name, service name, or brand name included in the read broadcast data is entered as a keyword into a search engine, and results displayed as advertisements in the search engine are excluded from the search results. Acquire one or more character strings described in any meta tag or title tag of the displayed website,
A generation device characterized by storing at least part of the one or more character strings, at least part of which does not include the keyword, in a searchable manner in association with the on-air data.

Images