WO2001067322A1

WO2001067322A1 - Information reception/transmission system

Info

Publication number: WO2001067322A1
Application number: PCT/JP2001/001722
Authority: WO
Inventors: Kuniaki Rokuto; Masahiro Kawano; Tsutomu Udagawa
Original assignee: Kuniaki Rokuto; Masahiro Kawano; Tsutomu Udagawa
Priority date: 2000-03-07
Filing date: 2001-03-06
Publication date: 2001-09-13
Also published as: AU3610301A

Abstract

A method and a system which enable individuals in the world to serve as information providers, can analyze information groups sent out from respective information providers and send appropriate information to appropriate information requestors, and can automatically decide valuable information to allow information appreciating persons to pay back to information transmitters. An information reception/transmission system comprises an information input means for inputting as event information information consisting of one of or a combination of characters, sounds and images of targeted events transmitted from the information terminal of an information transmitter via a prescribed information transmission route, and an information transmitting means for analyzing the above event information groups transmitted from an unspecified large number of information transmitters, and judging event information to be provided, for transmitting to information appreciating persons via the above information transmission route.

Description

Description Information reception / transmission system

Technical field

Regarding an information receiving and transmitting system that provides information transmitted from information providers (information transmitters) to information receivers, in particular, valuable information (from information groups transmitted from an unspecified number of information providers) An information receiving and transmitting system that automatically determines and transmits information from social articles with news properties to personal hobbies) and charges valuable information, and pays information usage fees from information recipients. The present invention relates to an information usage fee automatic billing method having a mechanism for paying back to a caller. Background art

Conventionally, when communicating information to an unspecified number of people, mass communication was used for commercial production activities using mass media such as newspapers and magazines, such as radio and television, as media for the mass media. On the other hand, they extracted from a large amount of information, reaffirmed the public's response, grasped the information as new articles and reports, and transmitted it to the public. On the other hand, dictionaries such as dictionaries have gathered immovable information based on universal facts, and have played a role in confirming reference with past common sense.

In addition, the Internet, which has been the subject of recent discussions, includes a search database engine (Database Engine), information disclosed by a content holder (Contents Holder), and personal (Personal User) and S OHO (Small Office Home Office). It focuses on the contents of the homepage, and is characterized by displaying the connection accurately. When the information receiver pays the information fee (royalties) to the information provider who provides the information, there is also a post-payment charge for the shareware (Share wa re) by download (Download), but the price exchange It is nothing but old-fashioned low-speed paperwork in the era of lending, where information speed is required as a performance as a means. There are also credit cards and other payment methods, but they do not add points in units of several monies, but require time and effort such as complicated authentication.

With the conventional technology described above, information can be transmitted to an unspecified number of people.However, it is possible to quickly obtain lively information on the spot, provide accurate information to those who require information, and provide that information. There was no obligation or mechanism to pay the service fees to the provider.

The present invention has been made in view of the circumstances described above, and the object of the present invention is to allow individuals all over the world to become information providers (information senders) even if they are not journalists. It is an object of the present invention to provide an information receiving and transmitting system and method capable of promptly obtaining useful information and providing appropriate information to required information seekers at a charge or free of charge. Another object of the present invention is to provide a system and method capable of providing a dictionary database such as a term (synonym) dictionary that is updated based on information from an information provider and is always kept up to date. In addition, based on a mechanism in which the information sender obtains the reward from the information receiver, it generates and consumes points that are the price of the information fee, and the information receiver automatically pays the information fee to the information provider. It is an object of the present invention to provide a system and an automatic information usage fee charging method. Disclosure of the invention

The information receiving / transmitting system according to the present invention includes any one of a character, a voice, and a video of an event of a category transmitted from the information terminal of the information sender, or a combination thereof. Information input means for inputting information consisting of a combination as event information via a predetermined information transmission path; and event information to be provided by analyzing the event information group transmitted from an unspecified number of information senders. And information transmission means for transmitting information to the information receiver via the information transmission path.Each individual in the world can be an information provider, The company is able to provide accurate or paid information to those who require information for a fee.

Further, the information usage fee automatic charging method according to the present invention is characterized in that information comprising any one of characters, voice, and video of an event of a class transmitted from the information terminal of the information sender, or a combination thereof is defined as event information. An information input step of inputting via the information transmission path, an information recording step of recording outgoing information created based on the event information, and a usage count of the event information to be charged. A billing information recording step of sequentially updating and recording each time the information is used, and information for paying the information usage fee from the information receiver receiving the event information to the information sender of the event information in accordance with the usage frequency It includes a charging step, and the information use fee can be paid back from the information receiver to the information sender for the information determined to be valuable information. BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 is a schematic diagram showing an example of the configuration of the entire system of the present invention. Fig. 2 is a flowchart showing an outline of the operation of the present invention. Figs. 3 (A) and (B) are examples of an event information input screen. Fig. 4 is a diagram for explaining the process for event information input from the information terminal, Fig. 5 is a flowchart for explaining the process of determining the term of natural language input, and Fig. 6 is a call Fig. 7 shows an example of the recording format of the recorded event information, and Fig. 7 shows the recording format of the components that make up the event information. FIG. 8 is a diagram showing an example of an expression, FIG. 8 is a diagram showing an example of the format of an information record registered in a synonym priority dictionary, and FIG. 9 is a flowchart for explaining a synonym priority determination process and a synonym priority dictionary update process. Fig. 10 is a diagram showing an example of the structure of the thesaurus priority dictionary database. Fig. 11 is a diagram for explaining the method of determining the information to be provided. Figs. 12 (A) and (B) Fig. 13 shows an example of the recording format of the adopted information source. Fig. 13 is a flowchart for explaining the context judgment processing and the update processing of the category-specific information context dictionary. Fig. 14 is the flowchart of Fig. 13. Fig. 15 (A)-(C) shows an example of an information browsing screen. Figs. 16 (A)-(C) shows another example of an information browsing screen. Fig. 17 is a diagram for explaining the flow of sending, providing and enjoying information. FIG. 18 is a partial diagram of FIG. 17, FIG. 19 is a flowchart for explaining the generation of the royalty and the payment process, and FIG. 20 is a diagram for explaining the process of creating the data used for the image recognition process. Fig. 21 is a diagram for explaining the recognition process of the components of the image, Fig. 22 is a flowchart for explaining the judgment process of the image information, and Fig. 23 is a storage form of the image information. Figure 24 shows an example of the contents of the image judgment dictionary, and Figure 25 shows an example of the judgment processing of images and audio information linked to information sentences. It is a flow chart. BEST MODE FOR CARRYING OUT THE INVENTION

In the present invention, for example, it is possible to know at the forefront the events of the genus (newly occurring lively events in the field with new nature, events that are notable even at the present time, etc.). The acquirer of the information allocates a dedicated ID (identifier that identifies the user) to those who want to use this mechanism, and transmits the information when it confirms the event of the attribute. And the information mediation system (contents holder [Contents Holder] etc.) The ID and the information are matched and collected, and the authenticity of the information is automatically judged from the time of occurrence of the new event, and from the context sent as information within a certain time (for example, 5 minutes), Judgment of term component matching is performed with a predetermined (eg, 60%) probability, terms with high weight are automatically evaluated and registered as categories, and information is acquired by the information acquirer (hereinafter referred to as information). Organizes information sent from the sender or the information provider).

Then, based on the organized information, event information that includes a context with a high frequency of occurrence of each term component and in which the frequency of occurrence exceeds a predetermined threshold is automatically judged as novel and reliable optimal information. And adopt it as valuable information and record it. The recorded information is transmitted in real time to the information terminal of the information receiver and provided, or provided by a browsing operation from the information receiver. Then, each time the information receiver uses the information, the usage frequency (point, int) of the event information is sequentially updated and recorded, and the information receiver who receives the event information sends the information to the sender of the information. Pay the information usage fee according to the above usage frequency. The person who receives the information usage fee is one or more persons who provided the valuable information described above, for example, according to a predetermined amount of points (for example, 100 points Z month). Then, the amount set by dividing the amount set by the number of providers of the information is provided to each information sender in the name of the information fee.

It also has a dictionary database that registers synonyms or similar image elements as similar elements for each similar element, analyzes similar elements included in the event information group transmitted from each information sender, and registers similar elements with a high frequency of occurrence. If the similar element to be registered is not registered, the similar element is automatically additionally registered with the frequency of occurrence described above in the dictionary, and the similar element to be registered is registered. If it is, the frequency of occurrence of the similar element in the dictionary database is automatically updated so that it is always up-to-date. A term (similar image element) dictionary that maintains the state of is provided.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In the embodiment described below, a case will be described as an example in which an information usage fee is paid to a person who first transmitted identical information for the same event. The information usage fee may be paid to all the people who sent the information. Also, an example will be described in which only the highest priority information and the information of the person who transmitted the information first are recorded in the information group for the same event.However, information having a somewhat higher priority is recorded and the information is recorded. It may be processed as a provision target (and billing target). Furthermore, information with low identity, that is, information with low credibility and low reliability is recorded so that it can be provided.For example, the degree and order of credibility are displayed on the information terminal side of the information receiver, and a selection instruction is given. When displaying the contents, it may be treated as a chargeable object.

FIG. 1 is a diagram showing a real-time information deduction to send back for automatic royalty point counting system according to the present invention, which is referred to as a “RAIDSARPS system” for convenience. 1) schematically shows an example of the overall configuration of FIG. The name of the information billing system in the RAIDSARPS system will be referred to as the "information billing system". In a broad sense, the "RAIDSARPS system" itself refers to the "information receiving and transmitting system (system of the present invention)". The present invention has a function to automatically collect and accumulate information such as events and hobbies of genera, and to automatically judge novelty (or attention) and reliable information in real time. It has a function to send information to the public and a function to automatically pay royalties (information use fee) from the information receiver to the information sender. In Fig. 1, 1 is the information terminal of the information provider (information sender) and 2 is the information terminal. The information terminal of the user, 10 is the information billing system that forms the basis of this system, 20 is the contents holder, 30 is the search engine, and 40 is the information transmission path.

The information terminals 1 and 2 connected to the information transmission path 40 are general-purpose information processing terminals having a communication function, such as a mobile phone such as a PHS (Personal Handy phone System) and a PDA (Personal Digital Assistants). In addition to portable information and communication equipment, portable or stationary general-purpose computers such as PCs (Personal Computers) and WSs (Work Stations), and car navigation systems, the following information processing equipment Is assumed. For example, information reproducing devices include large-sized display devices around stations and outdoor facilities, electric bulletin boards, electronic bulletin boards, nursing care assistance mechanisms, medical assistance devices, Internet display control devices, message display devices, and vehicles. Operation assistance device, voice synthesizer, portable radio display, wristwatch digital display panel, telephone display, music recording / playback device message, image from game equipment, voice display, ground wave BS ZC SZC AT Television equipment including digital and analog by V, system display for police and security services, electronic display of emergency response equipment, message board for parking guidance display, status indication for road operation assistance Equipment, road guidance display devices, vending machines for tangible and intangible objects, etc. Furthermore, it is assumed that these devices are equipped with information input means or connected to an information input device to be used as information terminals 1 and 2 having a function of transmitting and receiving information and receiving information.

The information transmission path 40 is a path for connecting information processing devices by wireless or wired (a combination of both) and transmitting information, and is an arbitrary communication line network such as the Internet or a communication network composed of paths. Network. In the example of Fig. 1, the communication network 'is provided by a circuit construction and maintenance company. It is composed of an Internet / packet communication network 41, a provider 42 or an exchange 43 provided by an infrastructure provider, an antenna information transmitting station 44, a satellite radio transmitting station 45, etc. A global network that connects terminals 1 and 2 (clients) and an information billing system 10, content holder 20, and search engine 30 provided by a web information provider by wireless or wired connection Is formed.

This RA IDSARPS system is equipped with a multifunctional server that has at least a gateway function for interconnecting information terminals / networks of different communication protocols and a database server function.Wireless or wired communication functions are not available. An information processing device that does not depend on a specific device. The information transmission path is configured by an information transmission route that does not depend on wireless or wired communication, and that does not depend on a specific communication protocol and a specific route. It enables information processing with various information processing devices via a multifunctional server. As a specific communication protocol, at least various protocols conforming to the open system interconnection (OSI) standard can be used. However, a protocol using an information record according to the present invention described later is applied to a higher-level application protocol related to transmission and reception of event information.

The information to be processed in the present invention indicates any information from social articles with news to personal interests, and does not limit the content of the information. Information to be obtained, in particular, information on the events of the genus (such as information on newly occurring news events at lively sites and information that is notable even if it is not at this time) is suitable. . Although such news-related information is usually provided by the press, in the present invention, the information provider that provides the information is an individual who transmits information using the information terminal 1. Yes, anyone can be an information provider. For example, for a newly occurring event, The acquirer (information provider) of the information that can be known confirms the event by himself, and the information consists of a combination of textual information, image information (still image or video image information including video), and audio information related to the event. (Hereinafter referred to as “event information”) from the information terminal 1. The "voice information" referred to in the system of the present invention refers to all audible sounds, such as human voices, object sounds, and music. In the following description, the information provider's voice and other audible sounds are used. For distinction, the former is referred to as audio information, and the latter as music information.

The information billing system (information receiving and transmitting system) 10 collects and records event information transmitted from the information terminal 1 of an unspecified number of information providers. Then, in the collection process, the context of the event information, the components such as images, etc. are analyzed by the information judgment algorithm described later, and the truth of the information (credibility and reliability), the novelty of the information, Judge in real time the identity of the information with the information provider. In the present invention, when it is determined that information is for a new event, for example, starting from the time of occurrence of the first report for that event, a reliable superiority is obtained from the group of information received within a certain time for the same event. Information is automatically selected, and the event information of the sender is stored in a database and recorded as information to be provided. In addition, the collected information is stored and recorded without making a judgment in real time, and whether or not the information is currently worthy of attention is determined based on a predetermined criterion. You may do it.

In a context determination algorithm that analyzes the context of character information, for example, a group of event information transmitted from each information provider is input, and the components of each event information (in this embodiment, each term (or word included in the context) )), And assigns high priority (weight) to the terms that occur frequently and sorts them by term. Information with the highest priority (priority) among the contexts composed of each term combination, In other words, the most frequently occurring terms in the context Information that contains (the terms that have the highest number of matches) and that has a priority value (frequency of occurrence) exceeding a predetermined reference value is determined and adopted as reliable and superior information. Further, in the present embodiment, for example, a limit is set for the information collection time, and the above-described reliable superior information is selected from the information group of the events of the genera that are simultaneously and frequently transmitted within a fixed time (for example, 5 minutes). Is automatically selected.

The event information adopted in this manner is stored and recorded on a data base of the information billing system 10 and distributed and provided in real time to the information terminal 2 of the information receiver, or provided by the information receiver. Provided on request from the information terminal 2. Each time the information receiver 2 accesses (views information) from the information terminal 2, the information amount (royalties) for the information is automatically accumulated as a frequency in the information billing system 10, and royalties are given to the information provider. You are getting paid. Payment of this loyalty is made by credit card @ electronic money, etc., and is automatically settled every predetermined period. In the example of Fig. 1, as shown in the statement of payment in Fig. 1, 100 yen is withdrawn from the account or electronic money as a communication fee to the information provider, and the loyalty of the information provided This shows an example where 15 yen was paid. On the other hand, an example is shown where a communication fee of 320 yen and an information enjoyment fee of 55 yen have been paid to information consumers.

In addition, in the system of the present invention, in addition to providing information having news properties in real time, a term dictionary that changes with the times is automatically created and provided. Priorities are given to terms (synonyms) that indicate the same thing, such as "I", "I", "I", "I", and "me", according to the frequency of occurrence of each term. A synonym priority dictionary is provided.Every time event information sent from each information provider occurs, it is categorized by synonym. The priority is updated to dynamically change the database of the thesaurus priority dictionary. In this way, the “synonym priority dictionary” dynamically changed according to the terms transmitted from each information provider is suitable for use as, for example, a buzzword dictionary or the latest term dictionary, in addition to being used in the context determination algorithm described above. For example, it is provided as valuable information to the content holder 20 and the search engine 30 in FIG.

The content holder 20 and the search engine 30 according to the present invention are general ones used in an information providing system using the Internet and the like. Combined by one protocol to form the RA I DSARPS system of the present invention. The information billing system 10, the content holder 20, and the search engine 30 are not limited to those having independent device configurations, but may be any type in which functions and hardware components are divided, integrated, and combined. Embodiments are included in the present invention.

With the above configuration, the flow of the entire processing of the information billing system according to the present invention will be described with reference to the flowchart in FIG. Note that the processes (steps S1 to S6) shown in ① to ① of Fig. 2 are described in the order of ①, ②, ⑤, ⑥, ④, and ③, and the processing of character information will be described first.

(1) Information input window processing for text, images, and audio via the Internet (Step S1)

First, an input form of event information input by the information provider from the information terminal 1 will be described. When information on the events of a genus, such as a newly occurring event, is input and conveyed in a sentence, the word components such as the subject, predicate, and modifier must be clear, the point must be clear, and more quickly It is preferable to be able to enter. Therefore, in the present invention, the free form is used as the input form of the event information. In addition to the form of input, a form in which only key points are input in a preset format is also adopted.

FIGS. 3 (A) and (B) show an example of an event information input screen displayed on the display unit of the information terminal 1 of the information provider, and FIG. 3 (A) shows an Internet connection. Fig. 3 (B) shows an example of a screen when inputting from a mobile phone with functions. Fig. 3 (B) shows a portable personal computer with a screen that is somewhat larger than a small information terminal such as a mobile phone. This is an example of a screen when inputting from an information terminal of this type. This example shows an example of a display screen of an Internet browser.

In this example, in the input form in which only the gist is input, as shown in the display examples in FIGS. 3A and 3B, "where" and "who (what)" In this way, words or word delimiters (particles, auxiliary verbs) by “de,” “ga,” or “do” are displayed in advance so that information on the ongoing event can be input quickly and simply. ing. Such delimiters such as “de,” “ga,” and “do” are used to increase the input speed of the information sender and to make the word components clearer and to speed up the analysis speed. is there. When inputting information from an information terminal having a relatively large screen, as shown in Fig. 3 (B), there are two types: one is to input only the essential points, and the other is to input in free sentences (natural language). You can choose. In the case of the latter input form using natural words, the expression “de” that specifies a place, and the subjects “ha” and “ga”, which are rearranged during the context judgment process, are input in which only the essential points are input. As with forms, words such as "where" are included at the beginning. As an actual input example, when transmitting information such as “A fire broke out in Shinjuku”, the input format shown in the input window of the input screens Gl and G2 in Fig. 3 will be used.

Although the input screen differs depending on the browser used for input, The obtained information is taken in as a series of sentences and sentence analysis is performed. Further, the transmission information targeted by the present invention is not limited to the text composed of the character strings as described above, but is also intended for information such as images, voices, music, and the like. Images, voice or music can also be transmitted. For example, in the case of the mobile phone shown in FIG. 2 (A), if a camera image can be captured in the Internet mode, "image present" is checked. The same applies when there is voice. When inputting from an Internet browser in the general-purpose video or the like shown in Fig. 3 (B), if there is an image or sound to be transmitted, the image is pasted via, for example, a clipboard. In addition, the audio and image files recorded on the recording medium by another device are transmitted by the control means on the information terminal side in accordance with the instruction of “with sound” and “with image” in FIGS. 3 (A) and 3 (B). It is good also as a form taken in.

The event information thus input is transmitted in response to a transmission instruction from the information terminal 1 and transmitted to the information billing system 10 via the information transmission path 40. The character information transmitted from the information terminal 1 is fetched as a series of sentences and sentence analysis is performed in any of the input forms.

② Sentence judgment processing for character information (step S 2)

Information processed by the information billing system includes text, image, and voice information. Here, the process of determining the components of a sentence input in natural language (step S2a) will be described.

In the form of natural language input, a participant of a particle or a verb is included. In the context analysis means of the information billing system 10, in the case of the form of natural language input, unnecessary parts (unnecessary parts for indicating the meaning of words) in the sentence are deleted by the following processing to make sense. Extract and extract terms or words. First, from the newly generated information group (C), all of the information sentences (C _Q , C, C ₂ , _... C _n ) included in the information group (C) The term or word Four

For (P), it is determined whether or not the part to be deleted exists in the context, those parts are removed, and all of them are keyworded. For example, one piece of information in the comma format “,” Statement P. , P i, P ₂ , ··· P _m .

The parts to be deleted are the following five kinds of parts in Japanese, for example, [J_A] to [J-E]. It is stored as a keyword used in processing or the like. <5 types to be deleted>

[J-I A] Prefix (such as "O" in "Story")

[J-I B] suffixes (such as "san" after the word "aunt")

[J-I C] conjunctions (such as "aga", "and", "but", "there"

)

[J-D] Particles ("we" in "we", "wo" in "you", "from town"

"Power", "no" etc.)

[J-I E]: Verb conjugation part (such as "to" or "to")

That is, other than the above [J-I A] to [J-E], adjectives such as "deep", adjectives such as "great", nouns such as "me", adverbs such as "really",

Adverbs such as "Kururu" and "this" and "that" in "Kururu Day", inspirational words such as "saya" and "maa", "no" in "come likely" and "unknown" Are treated as terms or words (P) without elimination.

Fig. 4 shows a specific example of a sentence input from an information terminal and an example of an information record that records the input sentence, the excerpt form of the term, and the combination pattern. In this example, a sentence (C., C or C ₂ ,... C „) input in the form of natural language input and a meaningful term or word (P., P, P ₂ ,. _m ) is extracted, and the information sentence format is extracted. You. [E0D] in FIG. 4 is an end mark indicating the end of data, and the extracted terms are stored in the information record as, for example, an array delimited by delimiters ",".

As described above, in addition to targeting outgoing information entered in Japanese, this system also accepts information on events of genus originated in languages of countries around the world. Even in the case of input in a language other than Japanese, in the case of natural language input, the context is analyzed and meaningful terms or words are extracted and converted into keywords, which are expressed in comma format “,”. One information sentence? . ,? ₁ ,? ₂ '? „Create ₁ automatically.

In this system, in order to be able to input in the natural language of each country in the world, the parts to be deleted are set for each language such as Japanese or English. "Specific proper noun dictionaries (databases)" are set up, and parts that are not to be deleted are set. Refer to these dictionaries to determine and extract meaningful terms or words. Like that. In the specific proper noun dictionary, words that include people, names, and hiragana, and words that have continuous words and specific meanings (idioms) that the words bring, are registered. We try to identify the connection of terms that should not be.

In the case of Japanese sentences, as described above, for example, for connection that is attached to words such as “ga” “ha” “te” “ni” “oh” “ha” “to” “is” “mas” Delete a word.

Foreign languages are generally easier to process than Japanese. For example, in English, the parts that are clearly identifiable, such as “to”, “so”, “that”, “from”, “for”, “by”, “who”, and “wherej” are deleted, and meaningful words are extracted. For the sake of judgment, uppercase letters are temporarily converted to lowercase letters. The parts to be deleted are, for example, the following [E-A] to [E-E] in English. There are five types of parts. As in Japanese, these parts are deleted, meaningful terms are extracted and converted into keywords, and one information sentence P is listed in comma format “,”. , Automatically create P, P _2, the · · · P _m.

<5 types to be deleted>

[E-A]: Article ("a", "an", "the")

[E-B]: Interrogative pronouns' relative pronouns ("what", "which", "who", "whom", "whose", "that")

[E-C]: 畐 'J ("so", how "," when "," why ")

[E—D]: Adjectives ("at", "on", "by", "from", "in", "into", "to", "for", "of", "with")

[E — E] conjunctions ("and", "but", "or", "because", "if", "as", "than")

Information sentence P created as described above. , P i, P ₂ ,..., P _m are stored together with information such as a user ID (User ID) for identifying the information sender and the date and time.

Here, the above-mentioned natural word input term determination processing in step S2a will be described with reference to a specific example with reference to the flowchart in FIG. 5, in which event information from the information provider's information terminal 1 is received. In the information billing system 10, original data such as character information is developed in a field of a transmission information recording area and stored together with information such as a user ID and a transmission time. Fig. 6 shows an example of the recording format of data of outgoing information (Protocol of data transfer). The outgoing information is one record (variable length). Original data, user ID, date and time, image information storage address (file name), audio (sound) information storage address (file name), priority, end mark (E0C: end Outgoing code) is recorded. In the present embodiment, of the information group for the same event transmitted simultaneously and frequently within a certain time, only one piece of information is finally recorded, and when there is a plurality of pieces of the same information, Finally, the user ID (character, voice, and image), date and time of the person who sent the information first are recorded in the corresponding fields in Fig. 6. In addition, the priority (Priority) in FIG. 6 is n = 1 if it is new, and if there is existing information, it is counted up (+1) (step S11).

The context analysis means analyzes the original data of the character information expanded in the transmission information recording area, and deletes the data in the deletable dictionary database (JDB1, EDB1) in which the parts to be deleted in the text are set. Based on the data in the deletable dictionary database (JDB2, EDB2) in which parts not to be deleted in the text are set, determine and extract meaningful terms (or words) (Step S 1 2). Then, event information C ′ i in which the extracted terms P i are arranged is created, and as shown in FIG. 7, an information record for each term P i is created, and each information record of the event is stored. As shown in FIG. 7, the information record of one term is composed of, for example, the term P i as a keyword, the information sender's user ID, date and time, and priority. The information records of the components (in this example, terms or words) constituting the information are linked and stored (step S13).

Next, the synonym priority judging process (step S2b in Fig. 2) in Fig. 2 will be described. The thesaurus priority determining means, each term of the extracted term _{(P) (P Q, PP} 2, · · · P m) with respect to, compared to the synonym that is registered in the "synonym priorities dictionary" The priority of this term (P) is increased to places where similar terms are registered, and the dictionary is updated or registered as new, and the dictionary is increased. Fig. 8 shows the format of the information record registered in the thesaurus priority dictionary. For each term P i, the information records of the synonym P j of the term are arranged in order of priority (in this example, the number of accesses). The configuration is as follows. As shown in FIG. 8, for example, each information record P (i, j) stores the user ID, date and time, and priority of the information sender using each term as a keyword. The synonym priority determining means updates the priority counter (= utilization rate count 1) indicating the number of occurrences of the term (P) (in this example, the priority in the information record is + 1) and updates the term. Record in P information record. Then, judging from the counter, the term (P) is rearranged in order of priority (priority). However, if the recorded user ID is a newly registered synonym, it will be the speaker (information sender) at the time of the new occurrence.

Here, the synonym priority judgment processing and the synonym priority dictionary update processing in step S2b described above will be described with reference to a specific example with reference to the flowchart of FIG.

The synonym priority determining means initializes the subscripts i and j of the array data P (i, j) of the new information P (step S21), and then newly transmits the new information P at the same time within a certain period of time. For the group P _m (i, j) (step S22), each term (word) included in the new information group P _m (i, j) is registered in the thesaurus priority dictionary data SPDD. Search for synonyms that exist. Figure 10 shows an example of the structure of the thesaurus priority dictionary database SPDD. In the thesaurus priority dictionary, each term (synonym) is used as a keyword, and information records of each synonym are linked in order of priority. And is recorded (step S23).

As a result of searching the thesaurus priority dictionary data SPDD, if there is an information record of a synonym that matches the term (word) component, the priority · Check the counter and update the counter. If there is no information record with a matching term, the term is set as a new term, and the priority is set to 1 and an information record is added (steps S24 and S25). Then, the process returns to step S22 and repeats the processing. If the processing of all new information groups P _m (i, j) is completed, the information records of the synonym group P _m are rearranged in order of priority (step S22). 26 6), the updated priority information (utilization rate counter = priority, counter) is recorded in the utilization rate counter record file CTD, and the sorted information records of the synonym group P _m are stored in the synonym priority dictionary. To reflect. At this time, if the value of the term determined as a new term is, for example, the value of the priority counter is equal to or more than a predetermined value (a term having a large number of occurrences), the information record is additionally registered as a new term in the thesaurus priority dictionary data SPDD. (Steps S27 and S28).

Next, the context determination process (step S2c in Fig. 2) in Fig. 2 will be described. Since each information sender transmits information in the same event in various sentences, the information group generated simultaneously and frequently within a predetermined time includes various terms or words (hereinafter, referred to as “words”). There is a sentence composed of a combination of Therefore, even in respect meaningful terms only the extracted information group (C), a respective information statement (:., The information source of C _l C ₂ ~ C _n (source), various terms In this system, in order to simplify this combination, the term at the top of the sorted (ordered) priorities by the processing in step S2b described above. Converting each sentence to (P.) has the effect of aggregating similar information sources.

As a result, of the information input and transmitted within a predetermined period of time (5 minutes in this example) after the information is generated, for example, each of the components that are character information components It is determined that the information with the most matching words contains the same meaning, the most people have the same expression for the information such as the news that has occurred, and is the most reliable information source. By doing so, it is possible to automatically verify the credibility.

Fig. 11 shows a specific example of a method for automatically determining credible information from a group of information corresponding to such events that occur frequently and determining the information to be provided. . In the context determination means in the information billing system 10, for example, each information (C _Q , C e, C ₂ ,...) Input and transmitted within a predetermined time (5 minutes in this example) after the information is generated. C _n ) are rearranged in descending order of the sum of the priorities (in this example, the sum of the occurrence frequencies of each term) of each term (word) component (P., P i P ₂ , P _m ) constituting the information, Information C of the highest rank (the one with the highest number of terms (synonyms) matching) among the information whose sum of priorities exceeds the recruitment threshold. Is determined as information to be provided. In this way, the person who sent the highest priority information (= the user with the highest probability (%) and the counter value indicating the frequency of the same number of accesses indicating the highest value) and the time series The person who sent the information (user ID) at the earliest time is recorded (= adopted) as this information source.

Fig. 12 (A) and (B) show specific examples of the recording format (Protocol) of the adopted information source. The format of the information record for recording the information source is shown in Fig. 6. The format is the same as the format of the information record developed in the transmitted information recording area. As shown in Fig. 12 (A), the information source P adopted. Is stored in an information record consisting of Keyword, Content, User ID, Date, Image Field, Audio Field, Priority, and Data End (EOC) fields. Key words and dates are used as keys. Recorded in the “Information Context Dictionary Database by Category” You. In this example, the information stored in each field of the information record is a keyword and ancillary information C, as shown in FIG. 12 (A). (The original text extracted at the highest frequency), the user ID indicating the sender of the text information, the date and time, the storage address of the image information (file name in this example), and the storage address of the audio (sound) information (the book In the example, the file name) and priority information. In the information record recorded in the “category-based information context dictionary database” described above, there may be multiple information sources of the same keyword. In this case, as shown in FIG. 12 (B), Multiple information records of the same keyword are recorded. For example, if only a keyword (“Shinjuku” in this example) is searched as a key, the event information of a plurality of categories (contents) of the keyword is obtained. Provided. The sender of the image information and the sender of the voice information may be different from the sender of the text information. In this example, the sender of the image information and the voice information will be included in the file name of the file that stores the audio data. By setting the user ID of the information sender in the, the information sender can be identified.

As described above, the information with the highest reliability and credibility is automatically extracted from the information on the event transmitted by a plurality of information senders and is adopted as the information to be provided to the information receiver. And recorded in the “Information Context Dictionary by Category”.

Here, the context determination process in step S2c and the process of updating the information context dictionary for each category will be described with reference to specific examples with reference to the flowcharts of FIGS. 13 and 14. I do.

In the information billing system 10, after performing a predetermined resident process (step S 31), a counter “n” for counting the number of event information is set to 0 (step S 32), and within a certain period of time, The following steps S33 to S39 are repeated for the new information group transmitted from the information terminal 1 of each information sender. Then, the degree of coincidence between the event information C _n of the information sender and other event information groups (combination sentences: C 0, C _x , C ₂ ,-) is obtained. First, by entering the event information I _n of the information callers, updates the counter evening one T indicating an elapsed time from the time point of generation of the new information (Step S 3 3), the event information I _n to configure each term component (P., P E, P _2, ... P _m) and extracted (step S 3 4), creates an event information C _n consisting of the extracted term component (step S 3 5) o

Subsequently, the content of the event information C _n of the information sender and other event information groups (C ₀ , C or C ₂ ") transmitted before that are collated, and, for example, each term component (P a The number of occurrences of 'P ^ P ^ Pm) (priority described above) is regarded as the weight (weight), and the degree of matching of the event information c _n of the information sender to the event information group up to now (identity of contents) In this example, the number of occurrences K of the same word or term combination that constitutes the event information c _n of the information sender is counted (K = K + 1), and the total number of combination sentences (C) is calculated. The ratio (%) divided by is defined as the degree of matching.

For example, in the example in step _S36 , the term component of the event information C _n of the information sender is P. , P! , P ₃ , P ₄ , P _Q (Shinjuku, Nishi-Shinjuku, around Shinjuku, or north of Shinjuku) +? (Fire, fire) + P ₂ (around, around) + P ₄ (ramen shop, soba shop) ) Indicates that 34% of the event information has occurred so far. This ratio (%) is recalculated for other event information (step S36).

Then, the maximum frequency in the combination sentence (C ₀ , C i, C ₂ ,) is compared with the threshold value (the threshold value to be used: 60% in this example) (step S37). The event information (combination sentence (C _Q ) with the highest frequency) is determined to have authenticity and novelty, and is set as information to be provided (step S38). Then, the elapsed time T is compared with the information collection time St, and a predetermined time In the example, it is determined whether or not the time has elapsed (St = 5 minutes) (step S39). If it has not elapsed, the process returns to step S33 and repeats the processing for the next event information.

Then, if a predetermined time has elapsed, the combination sentence (C) having the highest frequency is determined as the information to be provided. For example, Te example smell of the first 3 in the figure, ID: AFG002548887 01.31.2000-14: 26 "It's a fire in the vicinity of Shinjuku" that the information I _{_Q} (C _Q = Shinjuku, peripheral, fire) case is the beginning of the, 2000 Information sent within 5 minutes from 14:26 on January 31, 2014. Information I above. Followed by ID: AIS000361135 01.31.2000-14: 26 Information that “fire occurred around Shinjuku” ((^ = Shinjuku, around, fire, outbreak), ID: AT D007780125 01.31.2000- 14: Π "Is there a fire at the ramen shop on the north side of Shinjuku? J Information I ₂ (C ₂ = Shinjuku north side, ramen shop, fire, outbreak), ID: A GF029387265 01.31.2000-14: 27" Shinjuku is a fire. It is assumed that the information I ₃ (C ₃ = Shinjuku, fire, ramen shop, fire engine) was transmitted respectively. In that case, the context judgment algorithm by the context judgment means is “Shinjuku”. , “Shinjuku north side” is considered synonymous and weight (weight) 4, “fire” and “fire” are also considered synonyms, and weight 4, “surrounding” and “around” are also considered synonyms. Wait 2, “Ramen shop” and “occurrence” are weight 2 respectively. At the time of, in the sentence that combines terms, matching degree (probability) is adopted threshold = 6 0% or more of present such damage, the information C of the provision target is not set.

In the example in FIG. + And P + P ₂ of the information is assumed to have occurred in the 6 5% of probability, most began to disseminate information (C ₀₎ the row summer was a person of a user ID (AFG002548887), date and time (1.31.2000-14: Event information such as 26) is recorded in the category-specific information context dictionary database CTB. Therefore, the outgoing information is automatically organized and receives the first outgoing information. Five minutes later, for example, a message such as "11.31. 2000- 14:26 has occurred! Fire is happening around Shinjuku Membership number ..." Words with shallow weights (for example, “Ramen shop” and “3 fire trucks”) are judged to be of poor empirical value and are not used in this example. It may be recorded as “Ramen shop? 3 fire trucks?” And information can be transmitted as ancillary expression. Further, a threshold value may be provided in n stages, and a mark indicating the authenticity, a numerical value, or the like may be added according to the threshold value to transmit information.

Outgoing information C from the information billing system 10 is delivered in real time to the information receiver's information terminal 2 via the content holder 20 or the search engine 30 or directly from the information billing system 10 ( Or, it is transmitted by a request from the information terminal 2 side (FIG. 14: step S40). Next, the consistency between the transmission information C and the past information group accumulated and recorded in the category-specific information context dictionary data CDB is checked (step S41), and when it is determined that they do not match (there is novelty). Then, the usage rate counter 1 is set to 1 as a new registration target (steps S42, S43), and the information record in which the usage rate counter 1 (priority) is rewritten is referred to in the information context dictionary database for each power category. (Step S44). On the other hand, if the event information (combination of terms) matches in step S41 above, it is determined that the event information is the same, and the utilization rate counter of the existing event information is updated (+1). (Steps S45, S46), and records in the category-specific information context dictionary data CDB (in this example, the usage rate count in the information record) (step S44).

Next, prior to the processing of ③ and の in Fig. 2, the processing of ⑤ will be explained respectively.

方法 How to extract voice information of text “image” through the Internet (Step S5):

The selected information sources can be browsed as current or past information from news documents on the Internet or keywords that came up with similar analogy searches.

FIGS. 15 (A) to 15 (C) show examples of screens for browsing information from a mobile phone having an Internet connection function. FIG. 15 (A) is an example of a screen for browsing information distributed from a content holder 120 of a content company having the function of the present system, as shown in FIG. 15 (A). In addition, on the screen of the mobile phone, for example, information such as the date and time of occurrence of the information, the event information in characters, the member number of the information sender, and the agreement with everyone (the number of senders of the same information) are displayed. If audio information is present, "Audio playback" is displayed as shown in Fig. 15 (A), and the audio transmitted by the information sender is reproduced and output in response to the information receiver's reproduction instruction. You.

Fig. 15 (B) is an example of a screen in which information is browsed as a "news search" engine 30 by a content company equipped with the functions of this system. In this case, any keyword (example in the figure) Then, enter “Shibuya”) and give a search instruction. The “News Search” engine 30 will perform the search process, and the search results will be displayed on the screen. ,

Fig. 15 (C) shows an example of a screen in the form of news distribution via a banner or the like from the content holder 20 of a content company that has contracted with the information billing system 10 equipped with the functions of this system. In the example shown in the figure, a window ニュース for news display is provided in a part of the display screen of other content information, and a character string (newly generated information) is horizontally scrolled in the window. In this case, the latest information at that time is sequentially displayed. In this case, if voice information is present, “voice” is displayed as shown in FIG. 15 (C), and, for example, text information is scrolled. When this is instructed, the audio information of the news is output.

On the other hand, Fig. 16 (A) to (C) show the case of browsing information from an Internet browser on a portable or stationary general-purpose computer having a relatively large screen such as a personal computer or a workstation. The example of the screen of FIG. FIGS. 16 (A) to 16 (C) correspond to FIGS. 15 (A) to 15 (C), respectively. (A) A form for browsing information distributed from the content holder 20, (B) A form of browsing information as a "two use search" engine 30 from a content company, and (C) News distribution via a banner etc. from a content holder 20 of a content company contracted with an information billing system 10 Screen examples in different forms are shown. Here, the point that the display method is different from FIGS. 15 (A) to (C) will be described.

If the information terminal of the information recipient has a relatively large screen, it is effective to transmit information (reports, etc.) using both characters and images. The example in Fig. 16 (A) shows an example of distributing and displaying image information of a map with a fire mark at the place where a fire occurred. Further, the example of FIG. 16 (A) shows an example of a method of browsing image information according to an instruction from an information terminal of an information receiver. On the other hand, in the example of FIG. 16 (B), there are a plurality of event information items corresponding to the search key, and four of these event information items are displayed simultaneously on the screen. The image information displayed on the screen is the image information included in the event information sent by the information sender. For example, this system (information billing system 1) 0, content holder 20 and search engine 30 side), and the image information of the map obtained by searching with the keyword of the character information described above may be displayed. Note that the functions related to these image information are provided by the above-described mobile phone. It may be provided in a small information terminal such as the above.

Here, the flow of transmitting, providing, and enjoying information in this system will be described with reference to FIGS. 17 and 18. FIG.

As shown in FIG. 17, the content holder 20, the information billing system (provider) 10, and the search engine 30 are provided by, for example, an information provider operating a “Web site. In the description of the embodiment, a case has been described as an example in which the information billing system 10 performs all processes from the process of inputting information from the information terminal 1 to the process of providing the information terminal 2 to the information terminal 2. In the example, a form in which functions are dispersed is shown.

In the RAIDSARPS system according to the present invention, in terms of provision of functions (information processing means) and use means and roles, in the content holder 20, as described in the A1 frame in FIG. By providing the RIDSARPS system synonym, image, and sound judgment routines to a website operated by an independent company, the middleware of the application provided by the website is provided. The license is used as information provision by distributing data such as displaying a message in a format like a banner that scrolls only news that has some novelty. .

On the other hand, in the information billing system 10, as described in the frame B1 in Fig. 17, the parent company of the RIDSARPS system manages all the essence of the system and licenses it. In addition to providing information, it guides the flow of information between the provider and the beneficiary (beneficiary) of the information, and plays a role in managing the log (Log) for billing and traffic control of information. It also performs system version-up, system revision-up, system maintenance, and system management. On the other hand, as described in the description of the C1 box in Fig. 17, search engine 30 has a website operated by an independent company. On the other hand, by providing the RIDSARPS system with synonyms, judgment routines such as images and sounds, and new information such as news, the use of applications and linkages provided by the website will be promoted. In addition to the existing search engine, this RIDSARPS engine is used to display the screen of the provided system (for example, "Powered By RIDSARPS") and license it.

Regarding the provision of information input, the content holder 20 can configure the input information as an independent file, and performs, for example, an information input window processing of characters, images, and sounds through the Internet ( In FIG. 18, A 2), the information billing system 10 forms a main file of input information, and this company takes the initiative in collecting and managing information (FIG. 18, B 2).

Regarding the provision of information output, the Content Holder 20 provides the output from what is composed as an independent file and provides the relational [consolidation] through some distribution linkages (Fig. 18, A 3) In the information billing system 10, all information provision activities (real-time information transmission, distribution, etc.) are performed (Fig. 18, B3). In addition, the search engine 30 mainly provides the linkage of the output part related to information enjoyment, particularly the part that is not in real time, as a search engine having similarity (Fig. 18, C3).

Regarding the count processing related to the billing, for example, the content holder 20 can also perform the count processing independently when the information receiver browses, and a Daros contract is made (FIG. 18, A4). The information billing system 10 occurs when an information sender, such as context data or synonym data, after an input sentence is automatically analyzed, is sent by an information sender and viewed by an information receiver. Each counting process is performed (Fig. 18, B4). Also search In the engine 30, the counting process when browsing the information receiver is performed by the information billing system 10 (FIG. 18, C 4). Regarding the billing process, the information billing system 10 manages related histories (relational and log) by a priority counter (a utilization rate counter) used by both the information sender and the information receiver. Carry out work related to billing and dispensing (Fig. 18, B5).

発生 Process loyalty occurrence and payment (step S 6)

Next, the processing of step (S6) in FIG. 2 will be described. First, the mechanism of royalty generation will be described. In this system, the side that receives the information of the information sender (recipient) accumulates the information usage fee (royalties) as a direct point and provides a mechanism to automatically pay the information sender. ing.

In the case of event information with news properties, as described above, among the information on the events of the genus that are transmitted simultaneously and frequently for the same event within a certain period of time, the highest matching rate of the information component is the highest. Superior information is selected and adopted as the information to be provided. The information source (information source) adopted or the information collected by the original (proprietary) algorithm of the system

(The information recorded in the synonym priority dictionary database, the category-specific context dictionary database, and the image information data file described later) is treated as valuable information, and the information billing system 10 system uses the information. There is a billing mechanism (means) for sending back the royalty from the Internet user who wants to view the information to the original information provider (= sender).

The information sender and the information receiver are organically coupled via an information transmission path 40 such as the Internet, and can use the information database integrated by this system to accurately and interactively (interactively). Censored in By applying this system, it is possible to automatically realize the above-mentioned billing form as a price for using information. The information billing system 10 has a billing means for automatically and collectively paying the royalties charged (charged) by users on a monthly or yearly basis determined by the supervisor of the system. In addition, this facility can be replaced by an electronic money card called a Debit Card (Debt Card) under its own point system or public monetary standards. Also has. Hereinafter, the processing of the generation and payment of the royalty (step S6 in FIG. 2) will be described with reference to a specific example with reference to the flowchart in FIG.

The usage rate counter that indicates the amount of information used is, for example, the “priority” in the information sender's information record recorded on a recording medium such as the CDB database of the information context dictionary by category shown in FIG. (Priority) "field. As the initial value of the utilization rate counter, the number of accesses of the information sender (the number of information senders who transmitted the same information) is recorded in the processing of step S28 described above. In this example, this value is used. Shall be taken over. In the information billing system 10, the reference values m and y (set values indicating the duty frequency) according to the amount of information used are set for each monthly or annual payment form (step S51). The amount of the royalty is evaluated based on the value of the utilization counter 1 PC in the information record with reference to the judgment criterion table in which the standard values m and y are set.

For example, if the reference value of royalty is a monthly fee, the access number of information is "100, 20,000, 50,000," and a yearly fee is charged. The number of times of access to the information is “150 0 0 0, 4 .., "Are set in stages in accordance with the usage amount.

The data (original data, element data, and processed data) of the character information, audio information, and image information that were adopted as the information provision target were converted to the synonym priority dictionary database SPDD shown in Fig. 19, and the category-specific context. Dictionary database CDB, audio information file (source) SID o, image information data file (source) GID o, music information file (source) MID o, image judgment dictionary database GJD, image information file GID, image 4-bit compressed information file Recorded in each information record in GCD, each information record of information to be charged has a utilization counter PC. In this example, all eight types of information stored in the eight types of data (files) are charged.

As described in the accounting counting process described above (see the description of A4, B4, and C4 in FIG. 18), these utilization rate counter PCs use the information accounting system 10 (or the content holder 2). 0 or the search engine 30) increments the number of times of browsing (access to the information) at the time of browsing the information receiver or the information sender.

The information billing system 10 refers to the above-mentioned judgment criterion table every month and every year (step S52), and determines the value of the utilization rate counter 1 PC of the information and the reference value m (or y) of the royalty. After the comparison (step S53), if the value of the utilization counter PC exceeds the reference value m, a payout process is performed according to the amount of the corresponding royalty. This processing is performed, for example, for each of the above eight types of information, the value of the royalty of the user ID is added, and payment is made to the information sender of the user ID. In addition, the information senders of the text, audio, and video information that make up the event information are different. In such cases, information usage fees are paid to each information sender. For information receivers, a utilization rate counter is provided in a personal information record that records personal information, transfer form, and the like. The information usage fee is billed (step S54).

Payments (transfers) to information senders and bills (debits) to information receivers are made by users via settlement systems such as debit, card, credit card, and electronic money provided by banks and credit companies. (1) It is implemented online after authentication by inquiry / verification of ID, password, balance, etc. (steps S55, S56). The information sender (information provider) for the same event has been described as an example of one person (the person who sent the information first). However, in the case of paying to multiple information providers, the information sender is , The cumulative points are m% (for example, 30%) of the amount obtained by dividing the number of information recipients by the number of information providers. May be provided under the name of In addition, the billing process requires redemption in six months (for example, six months), so if half a year (six months) elapses, the customer will be reset if he or she does not accept the redemption. However, only the accumulation in six months (six months) can be redeemed. Therefore, the absolute amount does not become excessively large, and it is assumed that a new site on the web (Web) that specializes in further cashing from points will enter.

In the above-described embodiment, the method of analyzing and processing “character information” is mainly described. However, in the present system, “image information” or “voice information (including music)”, Alternatively, in addition to outputting the information of both parties to the information recipient's information terminal as it is, the information components (information components) of image information and audio information are analyzed, and each information component is converted to text information. Ingredients (for (Word or word), and performs information processing such as extracting and categorizing meaningful information. Then, the audio information data file (original) SID o, image information data file (original) GID o, music information file (original) MID o, image determination dictionary database GJD and image information data file shown in Fig. 19 GID, image 4-bit compressed information file Information recorded in each information record in GCD is provided for a fee.

Although the order of the flowchart in FIG. 2 is reversed, prior to the description of the image / sound determination process in FIG. 2 ③ before the description of the voice determination process, the following describes a specific example of the process of determining the image information. .

判断 Judgment of image information (Step S 4)

In the image information determining means, each image component of each image information (G ^ G ^ G-'Gn) included in the newly generated image information group (G) is included in the information group (G). The image is analyzed by the following procedures [A] to [G], the image components are determined by processing such as pattern recognition, and the image information to be adopted is determined and recorded.

[A] Decomposition of image signal components

[Red: red], G [Green: green] B [Blue: blue], respectively, and from the maximum 32 bits full color, (a) .1 bits for each light primary color Image processing as a contrast of 0 or 1 (contrast: Contrast). Furthermore, (b) 4 bits (luminance (I

: Intensity) = 1 6-bit color conversion and image processing as 4-bit contrast [contrast: Contrast]. However, the original image is of variable length and is stored separately. As a judgment material, (a) 1-bit data and (b) 4-bit data are created by superimposing newly divided images of each RGB bit. Fig. 20 shows the above 1-bit used for image recognition processing. The process of creating the data and 4-bit data is schematically shown in the figure of the RGB traffic light. As shown in the figure, the R, G, and B signals of the RGB signal IS are input to the 1-bit synthesis filter FL 1 and the 4-bit synthesis filter FL 2, respectively, and synthesized. OS (a) and 4-bit OS (b) are output.

[B] Image compression

Visual similarity to an image in which the color of the image has been reduced to 4-bit color, by combining the 4-bit color RGB prepared in advance and thinning out the color that can be visually judged to be the same color by a 4-bit color palette The speed of recognition of the same pattern is increased by compressing the same color or visual color into a uniform color.

[C] Setting the estimated probability of image recognition

In this example, the estimated probability of image recognition is 60% for the basic type. If the degree of convergence is poor in the judgment processing of identity, the basic type of the match (match rate) is reduced from 60% to about 30%, and analogy is applied.

[D] Data size setting for image recognition

The size of the data used for image recognition determination is, for example, identification of 100 × 100 pixels (minimum pixel unit: Pixel) (= dot [Dot] unit in this case) approximating the golden ratio Use de overnight.

[E] Judgment of image components

The image information determination means searches for constituents such as objects (buildings, intersections, etc.) that specify locations included in the image, and other objects and people (faces, whole images, etc.), and converts similar image elements into images. It is determined whether or not it exists. As a method of the judgment search, the entire compressed image of the 4-bit data OS (b) created by the above [A] and [B] is shown in FIG. 21 in the frame of the size set in the above [D]. As shown, for example, from top left to bottom right And search. At this time, the undivided portion at the right end or lower end is assigned to the reverse scale (right margin) from the end. Such uniform divided into certain intervals (equal parts _{_{Method: GD 2, GD 3, GD}} 4, GD 6, GD 7, GD 8, GD 9) that does not work only extracted for image evaluation by Especially, when judging by the method of extracting the frame from the part that seems to be the center of the next existing image and extracting the inside of the frame (center estimation method: GD _{1 Q} , GD _{1 15} GD _{1 2} ) In addition, the 1-bit data OS (a) is highly effective for its center estimation. This searches for what each component of the image is.

[F] Judgment of analysis feasibility by linear and nonlinear analysis

Depending on the image, the object that is the subject of the image (Object: Ob] 'ect) repeats a shape that is basically the same within 100 pixels in the vertical direction and 162 pixels in the horizontal direction, which approximates the golden ratio, and expands. The main components of the image can be changed freely, such as zooming, shrinking, inverting, changing the angle, or inverting the color. This is assumed. However, the probability of a simple shape as this pattern is very thin, and by repeating a certain pattern, every time the computer calculates, the sine (Sine) and cosine (Cosine) of the trigonometric function become Tangent) requires a huge amount of computational algorithms to determine in real time, and the FPU (Floating Point Unit) used for floating-point calculations and the internal cache memory of the CPU operate at high speed Is required. .

Therefore, considering the problem of comparison by database access, we examined the range that can be satisfied only by integer-type data overnight reference comparison, and the case that it was faster to make a judgment with some numerical operation. Testing] is required. In this case, the difference vector change algorithm of the light (light source · 'shadow' viewpoint) due to the weather, which can be fixed conditions, and the camera 'angle' There is a case where a photographing condition specified by the user is included. Therefore, taking into account the balance with the assumed ones, such as taking in simple figures and modifying specific colors, one that creates a database and one that solves it using a Fourier analysis or nonlinear fractal operation that can perform linear analysis, etc. It is necessary to determine whether it is data reference or calculation. In this example, it is determined whether or not to use the above-described analysis method for each part of the image. If it is determined that the analysis method cannot be used, the comparison operation is performed only by comparing the integer type data references.

C G] Determine the image (component) to be used

When determining and confirming the image (component) to be adopted, a 4-bit image recognition database is prepared in advance to find the comparison judgment of the image, and in the case of a new image, the image and the accompanying characters A new keyword is created from the information and automatically accumulates information for image analysis. Keywords are created in the same way as character information analysis, and keywords for database search are created. At this time, it is fully conceivable that the priority order of the text information and the information provider that is meaningful as an image may differ, so if there is an image component, after the text information is determined, the best candidate New keywords are matched to give priority to image information with high density.

Hereinafter, the image information determination process (step S4 in FIG. 2) will be described with reference to a specific example with reference to the flowchart in FIG.

The image information determining means repeats the following steps S61 to S76 for the image information group (G) transmitted from the information terminal 1 of each information sender within a fixed time, and each of images information _{(G i, G 2, G} 3 ... G n) contained within the (G) show, for each image component in the reference images components (identification data) and the image component Is determined. In this example, the image information to be processed (G i, G ₂ , G ₃ ... G _n ) is, as shown in FIG. 23, the file name shown in Image File 1 d in the information record described above. With image information File (source) Stored in GID 0.

In the image information judging means, first, after performing a predetermined resident process (step S61), 0 is set to a new information group processing counter "n" and a repetition count "i" of the repetition counter for the convergence process. Then, the count T indicating the elapsed time from the time when the new information is generated is updated (step S63) (step S63). Then, the image information group (G) is read from the image information data file (original) GID o, and each of the image information groups (G) included in the image information group (G) is decomposed by the method of decomposing the signal components of the image described in [A]. Each image determination component (1-bit data and 4-bit data) is extracted from the image information (G ^ G ^ G'.Gn) (step S64).

Subsequently, the image was compressed by using the image compression method of [B] described above to compress a visual similar color or a visually identical color into a unified color color (Step S65), and the image was compressed. The processing of the following steps S66 to S74 is repeated for each image, and image components are extracted, and the image component data is compared with the reference image component data for each image component. The similarity (coincidence) with the reference image component is analyzed to determine the degree of match (hereinafter referred to as “similarity estimation probability”). Here, the extraction and matching processing of the image components to be compared are performed in this example by dividing the image into equal parts and processing each divided block as an object (equal division method). There is a method of estimating and processing a block centered on the position (center estimation method). In this example, the similarity estimation probability is obtained using both. When calculating the above similarity estimation probability, first, each part of the compressed image is sequentially searched using identification data having a size approximated to the golden ratio, and when analyzing the constituent elements of the image, a linear It is determined whether analysis by nonlinear analysis is possible (step S66). If it is determined that the analysis is not possible, the similarity estimation probability is obtained by comparing only the comparison of the integer type data references. Therefore, if it is determined that the analysis is possible, the analysis is performed using one of Fourier analysis, fractal analysis, discrete cosine transform, difference vector analysis of the light source, viewpoint, and shadow, and collation calculation is performed for similarity estimation. The probabilities are obtained (step S67).

Then, assuming the adoption threshold (estimated probability of image recognition) Rate = 60% (step S68), each image component is searched by the above-described equalization method and center estimation method using the golden ratio block size. Then, the similarity between the image component and each reference image component is analyzed to determine the similarity estimation probability. Here, the image component data of each criterion described above is prepared in the image determination dictionary database GJD so as to be searchable by a keyword specifying the image component, and is searched by the keyword (Shinjuku, etc.) of the information record described above. The similarity is analyzed for the calculated image component data of each reference, and the similarity estimation probability is calculated.

FIG. 24 shows an example of reference image component data recorded in the database of the image determination dictionary. The reference image component data (lmage #) corresponding to each keyword such as face, circle, crossover, and Shinjuku is shown. l, Image # 2, Image # 3,...) are prepared in advance, and one of the similarity estimation probability by the equalization method and the similarity estimation probability by the center estimation method is used as the adoption threshold Rate ( If it is 60% or more, the image is judged to be an image corresponding to the keyword, and the priority counter 1 (= usage rate count 1) indicating the number of appearances of the image component is updated (step S69- S71).

In the above similarity (identity) determination process, if there is no similarity probability equal to or higher than the adoption threshold Rate (when the degree of convergence is poor), the adoption threshold Rate is gradually increased (for example, from 60%). (To about 30%), reduce the process, return to step S62, and repeat the process to apply analogy (steps S72, S73). Then, the processing of the image information is completed. If so, the above process is repeated for all the image information (G i, G ₂ , G _3, ..., G _n ) input so far for the next image information (step S74).

Then, for the image component (group) that is equal to or more than the adoption threshold Rate, the value of the utilization counter in the information record of the image component is updated, and the value is determined according to the hit amount (the value of the utilization counter in this example). Reorder the reference image component data group of the keyword and update the image judgment dictionary database GJD. At that time, as the latest information judgment image of the image information data file GID, it is joined with the keyword (P0) confirmation information of the category-specific information context dictionary database CDB and added as an extracted new image judgment dictionary database GJD (step S75).

Then, it is determined whether or not the information collection time St (5 minutes in this example) has elapsed (step S76). If not, the process returns to step S61 to return to the next new one. After the information is processed, if the time has elapsed, the image information is transmitted, for example, in association with the maximum frequency combination sentence (C.) (step S77).

As described above, in the character information determined to be provided, when it is determined that there is an image component corresponding to the keyword of the term (or word) component of the character information, the character information or the term (or word) component is added to the character information. At the same time, the image information is transmitted, and image components (groups) above the adoption threshold Rate in the new information group are automatically added (stored) to the image judgment dictionary database GJD. Create a learning database using information keywords.

Next, the determination (process S3) of the image / audio information linked to the information sentence (3) in FIG. 2 will be described with reference to the flowchart in FIG. ③ (Judgment of image and sound linked to information sentence: Step S 3) In this example, the analysis and determination processing of the voice information by the voice determination means is first performed after the determination of the character information is completed. The processing by the voice judgment means is the first-stage processing accompanying the analysis of the image and the voice. However, for image processing, those which could not be inferred from the transmitted information are handed over to the image processing means (image processing routine) in consideration of some real-time intervals. For those already found to match the character information, the dictionary structure for making more mature judgments on image data is strengthened. Hereinafter, the process of determining the image and audio information linked to the information sentence will be described with reference to the flowchart of FIG.

The voice judgment processing is performed when the categorization of the character information by the keyword is completed, and the image information and the voice information indicated by the address in the information record in the category information context dictionary database CDB are converted into the image information data. The file (original) GID o and the music information file (original) MID o are read out and subjected to the following processing (step S91). In the image processing unit, if the match with the image components is found, appending to the image determination dictionary database GJD with keys word (P _Q) is determined information and links Kategoryi specific information contexts dictionary database CDB (See step S75 above).

If a match with the image component is not found (step S92), or if there is no image information (step S93), a match between the key word of the character information and the audio component is found. It is determined whether or not it has been performed (step S94), and if a match with the voice component is found, it is linked to the category-specific information context dictionary database CDB keyword (P.) fixed information. And add it to the audio information data file SID o and record it. If no match with the audio component is found in step S94, it is determined whether or not a music file index exists (step S94). 5) If it exists, add it to the music information data file (original) MID o and record it. If these processes have been completed, or if there is no audio (Audio) information (step S966), the process proceeds to image information processing, and the process by the voice determination means ends.

Each of the above-mentioned means (context analysis means, synonym priority judgment means, context judgment means, billing processing means, image processing means, voice judgment means, etc.) uses the information record as input information, and It is realized as middle software (for example, a program in a routine format according to a predetermined input / output protocol), and a recording medium recording each step is provided to the content holder 20 or the like. In addition, the information display software licenses screen conditions for sending and receiving information to media manufacturers, or deploys it on relay software that constitutes an industry-standard simple man-machine-interface. It is realized by doing.

In the above-described embodiment, a case has been described as an example in which character information is mainly analyzed to determine the event information to be provided. However, the image information is converted into a similar image, which is a component of the image, by the above-described image processing means. The elements may be analyzed and processed in the same manner as the processing based on the term components to determine the event information to be employed, and the event information consisting of only the image information may be input. The same applies to voice information. Signal components may be processed in the same manner as the processing using term components (or voice-to-character conversion processing) to determine event information to be employed. Further, the order of each step mainly using the software described above is not limited to the order of each flowchart, and the configuration of each step group is not limited to the step configuration shown in each flowchart.

Further, in the above-described embodiment, the case where information having novelty is processed as valuable information has been mainly described, but the present invention is not limited to this. Instead, the information sent from the information provider (information sender) should be stored and recorded in a browsable manner, and the information requested by the information requester should be paid or provided free of charge. May be. In addition, information that can be captured as event information of a genus is automatically determined based on a predetermined criterion (for example, the number of times of browsing) from the stored and recorded information group, and the information is extracted and paid or paid for free. Information may be transmitted. In addition, basic information may be provided free of charge, such as shareware, and when all information is transmitted to the information seeker, the information receiver may be charged. Amount of money

The billing amount may be set based on (or the amount desired by the information seeker and the amount determined by the information provider) and charged according to the set amount. '

As described above, according to the present invention, information such as events and hobbies of a genus can be collected from individuals all over the world via a communication network such as the Internet, and is credible. Information can be automatically judged and transmitted in real time. In addition, for the information determined to be valuable information, the information receiver of the information can automatically pay the information provider for the information fee. Significance can be found in doing so.

In addition, similar elements with a high frequency of occurrence among similar elements (synonyms, similar image elements) transmitted from each information provider are automatically added to the dictionary database, or the amount of the frequency of occurrence is automatically updated. Therefore, it is possible to provide a dictionary of terms (similar image elements) that always maintain the latest state. Therefore, it can be suitably applied to the analysis processing of terms that compose character information, the analysis processing of individual image elements that compose image information, and the like. Can be provided to users via communication networks such as the Internet. Industrial applicability

According to the present invention, information such as events and hobbies of an attribute is collected from individuals all over the world via a communication network such as the Internet. Information that allows each individual to become an information provider (information sender), and to quickly obtain lively information on the spot and provide accurate information to required information seekers for a fee or free of charge. A receiving and transmitting system can be realized.

Further, according to the present invention, it is possible to realize an information billing system having a mechanism for paying back an information usage fee from an information receiver to an information sender. In addition, since the obtained information is analyzed to determine new and reliable information automatically, various types of information transmission that can transmit new and reliable information in real time It will be possible to realize systems (news systems for news and latest information from around the world, emergency information broadcasting (broadcasting) systems, etc.).

Further, according to the present invention, similar elements having a high frequency of occurrence among similar elements transmitted from each information provider are automatically added to the dictionary database, or the amount of the frequency of occurrence is automatically updated. Therefore, it is possible to provide a term (similar image element) dictionary that always maintains the latest state. Therefore, it can be suitably applied to the analysis processing of terms forming character information and the analysis processing of individual image elements forming image information. In addition, a buzzword dictionary and a database of the latest term (similar image) dictionary It can be provided to users via a communication network such as the Internet.

Claims

The scope of the claims

1. Information ^ Information input that inputs information consisting of characters, voices, and / or video of the event of the event of the class transmitted from the information terminal of the believer as event information via a predetermined information transmission path. Information transmitting means for analyzing the event information group transmitted from an unspecified number of information transmitters, determining the event information to be provided, and transmitting the information to the information receiver via the information transmission path; And an information receiving and transmitting system.

2. The information receiving and transmitting system according to claim 1, wherein event information transmitted from an information terminal of said information transmitter is transmitted in real time by said information transmitting means.

3. The information receiving and transmitting system according to claim 1, wherein the information terminal connected to the information transmission path includes a portable information communication device, a portable or stationary general-purpose computer, and an automobile navigation device.

4. The information reproducing device of the event information is a large display device around the station and outdoor facility equipment, an electric bulletin board, an electronic bulletin board, a care assistance auxiliary mechanism device, a medical auxiliary device, an Internet display control device, a message display device. , Vehicle operation assistance device, voice synthesizer, portable radio display, wristwatch digital display panel, telephone display, message recording / playback device message, game device image / voice display, terrestrial / Television equipment including digital and analog by BSZCSZCATV, system display for police and security services, electronic display of crisis countermeasure equipment, message board for parking guidance display, status grasp for road operation assistance Claims that include display devices, road guidance display devices, tangible and intangible vending equipment Range information transceiving system of Claim 1 wherein.

5. The information transmission path is a path for transmitting information by connecting the information processing devices wirelessly or by wire or by a combination of both. — The information receiving and transmitting system according to claim 1, comprising a communication network including an arbitrary communication line network and a route including a network.

6. The information receiving / transmitting system includes a multifunctional server having a gateway function for interconnecting information terminals Z networks of at least different communication protocols and a function as a database server. An information terminal having a communication function connected to an entrance / exit of a transmission path is an information processing apparatus independent of a specific transmission medium, a specific communication protocol and a specific device, and the information transmission path is wireless or wired. It is composed of an information transmission path that does not depend on a specific communication protocol and a specific path, and enables information processing with each information processing device via the multi-function server. The information receiving and transmitting system according to claim 1.

7. The information receiving and transmitting system according to claim 1, wherein the event information transmitted from the information terminal of the information transmitter is any information from a social article with news to an individual's hobby.

8. The event information transmitted from the information terminal of the information sender is information on a newly occurring event. Of the event information group transmitted from the unspecified number of information senders, the event information is The range of requests that automatically determines event information with a high frequency of occurrence of textual information, audio information, or video information components as new and reliable event information to be provided optimally. Information receiving and transmitting system described in 1.

9. The information transmitting means, among the event information group for the same event transmitted from each of the information terminals simultaneously and frequently within a certain period of time, transmits at least information having a high degree of identity of the information content to the event to be provided. The information receiving and transmitting system according to claim 1, wherein the information is transmitted.

10. The information transmitting means is the same as the information content of the event information group for the same event transmitted from each of the information terminals simultaneously and frequently within a certain time. 2. The information receiving and transmitting system according to claim 1, wherein information from a plurality of said information transmitters including information having a low degree of uniqueness is transmitted to said information receivers.

11 1. Term components are extracted from the context of the character information included in the event information, and term components that are the same or similar to the extracted term components are classified as synonyms for each synonym, and the frequency of occurrence of the synonyms And a context analysis means for analyzing the similarity of each event information by using the weight as a weight, and the frequency of occurrence of each term component in the event information group having the similarity and simultaneously and frequently transmitted within a predetermined time. 2. The method according to claim 1, wherein event information that includes a context of high frequency and in which the frequency of occurrence exceeds a predetermined threshold value is automatically determined to be novel and reliable optimal information. Information receiving and sending system.

1 2. Analyzing the dictionary database in which synonyms or similar image elements are registered as similar elements for each similar element, and analyzing the similar elements included in the event information group transmitted from each information sender, the similar elements with a high frequency of occurrence are analyzed. If the similar element to be registered is not registered as a registration target, the similarity element is automatically additionally registered by adding the occurrence frequency to the dictionary database, and the similar element to be registered is registered. 2. The information receiving and transmitting system according to claim 1, further comprising: a dictionary database updating means for automatically updating the frequency of occurrence of the similar element in the dictionary database if the information has been registered.

1 3. Information recording means for recording transmission information created based on the event information, and charging information for sequentially updating and recording the number of usages of the event information to be charged each time there is use by each information receiver. 2. The information processing apparatus according to claim 1, further comprising: a recording unit; and an information billing unit configured to pay an information usage fee from an information receiver receiving the event information to an information sender of the event information in accordance with the usage frequency. Information receiving and sending system.

1 4. Recognize the event information determined as the optimal information as valuable information 9. The information receiving and transmitting system according to claim 8, further comprising valuable information recognizing means.

15. The information receiving and transmitting system according to claim 8, further comprising employment information determining means for employing the event information determined as the optimum information as event information to be charged to an information receiver.

16. Charging information recording means for sequentially updating and recording the frequency of use of the event information to be charged each time the information receiver uses the information, and information of the event information from the information receiver receiving the event information. 2. The information receiving and transmitting system according to claim 1, further comprising: an information charging means for paying an information usage fee to a sender in accordance with the number of uses.

17. Information that is input from the information sender's information terminal and consists of any of the characters, voice, and video of the event of the class, or a combination thereof as event information via a predetermined information transmission path An input step, an information recording step for recording outgoing information created based on the event information, and a usage frequency of the event information to be charged is updated and recorded each time the information receiver uses the information. Information comprising: a charging information recording step; and an information charging step of paying an information usage fee from an information receiver who has received the event information to an information sender of the event information in accordance with the usage frequency. Automatic usage charge method.

18. Term components are extracted from the context of the character information included in the event information, and term components that are the same as or similar to the extracted term components are classified as synonyms for each synonym, and the generation of the synonyms is performed. A context analysis step of analyzing the similarity of each event information by using the frequency as a weight; and a frequency of occurrence of each term component in the event information group having the similarity and simultaneously and frequently transmitted within a predetermined time. Event information that includes a high context and the occurrence frequency amount of which exceeds a predetermined threshold is automatically determined to be novel and reliable optimal information, and 18. The information usage fee automatic charging method according to claim 17, further comprising an employment information judging step adopted as event information.

19. The information billing system includes a multi-function server having a gateway function and a database server function for interconnecting information terminals Z networks of at least different communication protocols. An information terminal having a communication function connected to an entrance / exit is an information processing device that does not depend on a specific transmission medium, a specific communication protocol, and a specific device, and the information transmission path depends on wireless or wired. And a data communication path independent of a specific communication protocol and a specific path, and enabling information processing with each information processing device via the multi-function server. The information usage fee automatic billing method described in item 17 of the scope.

20. —In the event information group transmitted simultaneously and frequently from each of the information terminals within a fixed time, the information sender who first transmitted the same information content for the same event is the information sender. 18. The information usage fee automatic charging method according to claim 17, wherein charging is performed as a payment target of the usage fee.

2 1. In the event information group transmitted from each of the information terminals simultaneously and frequently within a certain period of time, each information sender who has transmitted the same information content for the same event is charged to the information usage fee. The information usage fee automatic charging method according to claim 17, wherein the payment processing is performed as a payment object.

2 2 .—In the event information group for the same event transmitted from each of the information terminals simultaneously and frequently within a fixed time, at least the information with high similarity of information contents is targeted to the information receiver. The information usage fee automatic billing method described in Item 17 of the scope of the request to be respected.

23. In the event information group for the same event transmitted from each of the information terminals simultaneously and frequently within a certain period of time, information from a plurality of the information senders, including information with low information content, is collected. Send to the information recipient The method for automatically billing information usage fees according to claim 17, wherein the method comprises:

24. Equipped with a dictionary database that registers synonyms or similar image elements as similar elements for each similar element, analyzes the similar elements included in the event information group sent from each information sender, When the similar element to be registered is not registered as an element to be registered, the similar element is automatically additionally registered with the occurrence frequency in the dictionary database, and the similar element to be registered is 18. The method according to claim 17, further comprising a dictionary database update step for automatically updating the frequency of occurrence of said similar element in said dictionary database when registered. .

25. In the information charging step, when the information senders of the character, voice, and video information constituting the event information are different, the information usage fee is charged to each information sender according to the usage frequency. The information usage fee automatic billing method according to claim 17, which is to be paid for.