JP2003108566A - Information retrieving method and information retrieving device using agent - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information retrieving method and an information retrieving device using an agent capable of easily obtaining information from a wide range information source without requiring time and labor by autonomously retrieving the information on the basis of simple indication of a user. SOLUTION: This information retrieving method using the agent presents a plurality of predetermined agents for individually setting a proper information retrieving condition (102), and selects the agents by the user desired information retrieving condition from the plurality of agents (103), and retrieves user desired information (111).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information retrieving method and an information retrieving apparatus using an agent for retrieving information in a large-scale storage device in a device or information in a large-scale storage device in a remote place by a communication means. Regarding

[0002]

2. Description of the Related Art Information retrieval devices using agents are
For example, there is one described in JP-A-9-292895.

FIG. 30 is a block diagram of a conventional information search device using an agent.

The information retrieval apparatus using the conventional agent includes a voice input means 3001, a means 3003 for recognizing an input voice, a reliability determination means 3004 for a voice recognition result,
Data storage means 3007 for accumulating character screen data having a plurality of different facial expressions, screen output means 3013, and CPU 3006 for controlling these means are provided, and character screen data corresponding to the reliability level of the voice recognition result is stored as data. It is configured to read from the storage means 3007 under the control of the CPU 3006 and output from the screen output means.

[0005]

However, in the above-mentioned conventional information retrieval, the user has to set detailed information retrieval conditions every time information retrieval is carried out, which has a problem that it takes time and labor. It was

The present invention has been made in order to solve the above-mentioned conventional problems, and autonomously searches for information based on a simple instruction from a user, so that a wide range of information can be easily obtained without time and effort. It is an object to provide an information search method and an information search device using an agent that can obtain corresponding information from a source.

[0007]

An information retrieval method using an agent of the present invention presents a plurality of predetermined agents in which unique information retrieval conditions are individually set, and a user is selected from the plurality of agents. The agent is selected according to the information search condition desired by the user, and the information search desired by the user is performed.

With this method, the user can obtain desired information without spending time and effort.

Further, the unique information retrieval condition is presented sequentially by voice or in a list form on the screen according to a voice from a user or an instruction from a remote controller.

By this method, the user can surely confirm that the selected agent has the information retrieval condition desired by the user.

Further, when additionally registering the information search condition desired by the user, the information search condition desired by the user should be incorporated into the existing information search condition and added to the existing information search condition. If the same thing as the above exists, the additional registration is not performed.

By this method, the user can surely retrieve the information desired to be retrieved.

Further, it is checked whether the information search condition additionally registered by the user is semantically inconsistent with the existing information search condition, and if the information is inconsistent, the registration is rejected, or which one is to be registered. I decided to decide.

By this method, the user can easily determine that the selected agent was appropriate.

Further, when additionally registering the information retrieval condition desired by the user, the voice dialogue between the user and the agent is used.

With this method, the user can reduce the burden of additional registration.

Further, when additionally registering the information search condition desired by the user, the information search condition desired by the user is input in a natural sentence, and the sentence is analyzed to extract a condition necessary for the information search. It is decided to add to the table of the unique information retrieval conditions.

By this method, the user can save the trouble and time to think about the keyword.

Further, the agent for interacting with the user and performing a search according to the information search condition and the agent for reading the information by the information search are different from each other.

By this method, the user's interest in the agent can be maintained.

Further, when the sentence of the information obtained by the information retrieval matches the sentence existing in the agent in advance, the impression corresponding to the matched content is expressed to the user.

By this method, the user's expectation for the search result can be increased.

Further, the utterance content and utterance strategy of the dialogue with the user are determined according to the current emotional state of the user estimated from the history of the latest dialogue with the user.

By this method, it is possible to generate a dialogue that exactly matches the user's request.

Further, the utterance content and the utterance strategy during the dialogue with the user are determined according to the emotional state of the user estimated from the state of the user's voice.

With this method, it is possible to generate a dialogue that exactly matches the user's request.

Further, the utterance content and the utterance strategy during the dialogue with the user are determined according to the emotional state of the user estimated from the facial expression of the user.

With this method, it is possible to generate a dialogue that exactly matches the user's request.

Further, the utterance content and utterance strategy at the time of the dialogue with the user are determined according to the emotion history of the user comprehensively estimated from the dialogue history with the user and the state of the user's voice and facial expression. I decided to decide.

By this method, it is possible to generate a dialogue that exactly matches the user's request.

Further, a plurality of predetermined information sources are always searched using the information search condition of the agent, and the search result is reported to the user at regular intervals.

This method can save the user the trouble of calling the agent.

In addition, a plurality of predetermined information sources are constantly searched using the information search condition possessed by the agent, and when the search result reaches a preset number, the user is notified of the search result. I decided.

This method can save the user the trouble of calling the agent.

When a plurality of predetermined information sources are constantly searched using the information search condition of the agent and the total amount of data obtained as a result of the search reaches a preset limit capacity, the user It was decided to report the results of the above search to.

This method can save the user the trouble of calling the agent.

Further, when the information obtained by the information retrieval is stored in the storage area, the similarity with the previously stored information is judged, and when the similar information exists, the existence of the similar information is checked. I decided to report it to the user.

By this method, the user's familiarity with the agent can be increased.

When the information obtained by the information retrieval is stored in the storage area, the obtained information and the past information are analyzed and abstracted as a set of mutually associated keywords. It is generated and used for similarity judgment, and further, the past information is deleted after a certain period of time.

By this method, it is possible to save the memory of the retrieval result holding area and to compare the past and present retrieval results at high speed.

Further, when the information obtained by the information retrieval is stored in the storage area, the obtained information and the past information are analyzed, and a set of superordinate conceptual words of the extracted keyword is generated and similar. It is used for sex determination, and the past information is deleted after a certain period of time.

By this method, it is possible to save the memory of the search result holding area and to automatically determine the similarity regardless of the slight difference in the keywords.

Further, when the information obtained by the information retrieval is stored in the storage area, a set of keywords of both the new retrieval result and the accumulated past retrieval result is examined to find a pair of verb and object. Was found, and it was decided to determine whether or not the pair of the above verb and object match each other as a criterion for similarity.

By this method, the matching judgment can be speeded up.

Further, when new information is obtained by the information search, the newly searched information is analyzed to extract the theme that is the focus of the information, and the similarity judgment with the accumulated past information is judged. Decided to use.

By this method, it is possible to automatically determine the similarity regardless of the slight difference in the keywords.

The information retrieval apparatus using the agent of the present invention includes means for holding a plurality of predetermined agents, means for storing information retrieval conditions unique to each of the plurality of agents, and the agent. A configuration is provided that includes means for displaying on the screen and means for selecting the displayed agent.

With this configuration, the user can obtain desired information without spending time and effort.

Further, means for interpreting an instruction from the user input via a microphone or a remote control receiver, and means for reading out information retrieval conditions peculiar to the agent by a recorded voice or a synthesized voice according to the result of the interpretation. , A means for displaying a list of information retrieval conditions peculiar to the agent on the screen, and a means for selectively executing the reading means and the means for displaying the list or simultaneously executing both.

With this configuration, the user can confirm that the selected agent definitely has the same interest as himself.

Further, the means for extracting the information search condition from the input from the microphone or the remote control receiver and the information search condition stored in the storage means and the input information search condition are not the same. Means for determining whether the same thing exists, means for generating a sentence that reports that the same thing has been removed when the same thing exists, and means for outputting the sentence in synthetic speech ,
And a means for displaying the sentence on the screen.

With this configuration, the user can surely retrieve the information he / she wants to retrieve.

Further, a meaning understanding means for understanding the meaning of the information search condition to be additionally registered by the user, a meaning of the information search condition already stored in the storage means in the agent, and a new additional registration Means for determining a contradiction with the meaning of the input information search condition; and means for rejecting additional registration from the user and reporting the refusal of the additional registration to the user when the contradiction is detected, or the additional registration It is configured to include means for inquiring the user which one of the information search condition to be selected and the stored information search condition to be selected, and re-registering it in the storage area of the information search condition.

With this configuration, the user can easily determine that the selected agent is appropriate.

Further, means for understanding the meaning of the utterance of the user by using the voice recognition means and the language analysis means including the language dictionary, and whether the information retrieval condition desired by the user is included in the utterance of the user. It is configured to include a means for judging, a means for generating a dialogue strategy based on slot filling for drawing out information search conditions from the user, and a means for generating a synthetic voice according to the dialogue strategy.

With this configuration, the user can reduce the burden of additional registration.

Also, the means for inputting a sentence including the information retrieval condition desired by the user, the means for analyzing the sentence by comparing it with a typical sentence template, and the template for the typical sentence do not match. The input is analyzed using a language dictionary and a language analysis algorithm, and a means for extracting the information search condition desired by the user from the result of the analysis.

With this configuration, the user can save the trouble and time for considering the keyword.

The agent, which interacts with the user and performs a search in accordance with an internal information search condition,
A configuration is provided that includes means for generating a strategy for properly using the agent that reads out information by the search, and means for operating the agent according to the strategy.

With this configuration, the user's interest in the agent can be maintained.

Also, means for performing language processing of the retrieved information, means for understanding the meaning from the result of the language processing, and means for expressing the theme of the information retrieved from the meaning as a network of keywords. , Means for storing common sense peculiar to the agent in the keyword network, means for determining mutual intimacy of the keyword network, and means for generating an impression sentence corresponding to the result of the determination, And a means for synthesizing the impression sentence.

With this configuration, the user's expectation for the search result can be increased.

Further, means for monitoring the dialogue with the user and recording the dialogue history, means for estimating the current emotion of the user from the dialogue history, and a dialogue strategy for leading the emotion of the user to pleasure. A configuration is provided that includes means for generating, utterances are assembled according to the strategy, and emotional expression of synthetic voice and means for changing the facial expression of the character when the character is displayed on the screen.

With this configuration, it is possible to generate a dialogue that exactly matches the user's request.

A means for analyzing the prosody of the user's voice, a means for estimating the current emotion of the user from the prosody and the utterance content of the voice, and a dialogue strategy for leading the emotion of the user to pleasure. A configuration is provided that includes means for generating the speech, assembling the utterance according to the dialogue strategy, and means for changing the emotional expression of the synthetic voice and the facial expression of the character when the character is displayed on the screen.

With this configuration, it is possible to accurately generate a dialogue that matches the user's request.

Further, means for recognizing the facial expression of the user, means for estimating the emotional state of the user from the facial expression,
Means for generating a dialogue strategy that leads the user's emotions to joy; means for assembling a utterance in accordance with the dialogue strategy, and means for changing the facial expression of the character when the character is displayed on the screen and the emotional expression of synthetic voice. Prepared for configuration.

With this configuration, it is possible to generate a dialogue that exactly matches the user's request.

Further, means for monitoring the dialogue with the user and recording the dialogue history, means for analyzing the prosody of the voice of the user, means for recognizing the facial expression of the user, the dialogue history, and the prosody. A means for comprehensively estimating the current emotion of the user from the content of the utterance and a combination of facial expressions of the user; a means for generating a dialogue strategy for leading the emotion of the user to happiness; In addition to assembling a utterance, the emotional expression of the synthetic voice and means for changing the facial expression of the character when the character is displayed on the screen are provided.

With this configuration, it is possible to accurately generate a dialogue that matches the user's request.

Further, means for searching while switching a plurality of information sources in accordance with the information search condition inherent in the agent, and means for storing information on the result of the search,
And a means for reporting the result of the search each time a predetermined fixed time elapses or a predetermined time is reached.

With this configuration, it is possible to save the user the trouble of calling the agent.

Further, means for searching while switching a plurality of information sources according to the information search condition inherent in the agent, and means for storing information on the result of the search,
And a means for reporting the search result when the stored number reaches a preset number.

With this configuration, it is possible to save the user the trouble of calling the agent.

Further, means for searching while switching a plurality of information sources according to the information search condition inherent in the agent, and means for storing information on the result of the search,
It is configured to include means for calculating the stored total amount of data and means for reporting the result of the search when the total amount of data reaches a preset upper limit value.

With this configuration, it is possible to save the user the trouble of calling the agent.

Further, the means for storing the information obtained by the information search, the means for judging the similarity between the information of the past search results stored in the storage means, and the information for judging the similarity are stored in the user. And a means for reporting to.

With this configuration, it is possible to increase the familiarity of the user with the agent.

Further, means for storing information obtained by information retrieval, means for analyzing the information obtained by the retrieval and abstracting as a set of mutually associated keywords, the set of keywords and original text And a means for generating an interrupt every predetermined time, and a means for deleting the original text of the information that has been reported at the time of the interrupt.

With this configuration, it is possible to save the memory in the search result holding area and to compare the past and present search results at high speed.

Further, means for storing the information obtained by the information retrieval, means for analyzing the information obtained by the retrieval and abstracting it as a set of mutually associated keywords, and a superordinate concept of the keywords Means for determining,
A storage unit that stores both the superordinate concept and the original text of the search result, a unit that causes an interrupt every predetermined time, and a unit that deletes the original text of the information that has been reported at the time of the interrupt. Prepared for configuration.

With this configuration, it is possible to automatically determine whether or not there is a match, regardless of slight differences in keywords.

When the information is obtained by the information search, the verb and the object are stored as a pair in the set of both keywords in order to judge the similarity with the accumulated information of the past information search results. And a determining means for determining whether or not the verb and object pair match.

With this configuration, the similarity judgment can be speeded up.

Further, when the information is obtained by the information retrieval, a means for analyzing the retrieval result obtained by the information retrieval in order to judge the similarity with the information of the accumulated past retrieval results, and the analysis. And a means for storing the theme together with the original text of the information which is the search result, and a means for judging whether or not the theme matches. And

With this configuration, the similarity can be automatically determined regardless of the slight difference in the keywords.

[0087]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to FIGS. 1 to 29.

FIG. 1 shows a flow chart of an information retrieval method using an agent according to the first embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, in the information retrieval method using this agent, a list of images of a plurality of agents acting on behalf of the information retrieval work and their characteristics is presented for the user (102), and the user explains the images and the characteristics. Select an agent who is likely to have a tendency similar to one's interest from (103). For example, in the case of "Oyaji Agent," an image of an old man who seems to be stubborn and characteristics such as "Slightly stubborn. I am interested in political economy. Weak to grandchildren and wives." If you want to request information such as "Father's Agent", you can expect that you can get almost the information you want.

Then, the user selects the name or number. Only the selected agents remain, the other agents are erased. As described above, each agent has a unique information search keyword (information search condition). For example, in the case of a business man agent, words such as “salary”, “business”, “selling item” are preset as information search keywords. When the user requests the agent to present information retrieval keywords specific to the agent, a list of keywords is displayed on the display. Depending on the case, it is also effective to read the registered contents aloud with a synthetic voice or the like.

When the agent is selected, the information search operation of the agent is started (111). The user
When the user thinks that the presented information search keyword is insufficient, the user can additionally register the information search keyword.

At this time, it is checked whether or not the keyword overlaps or is inconsistent with the one registered in the past (10
7). If there is the same keyword in the past, the registration is canceled (108) and it is checked whether the meaning conflicts (1).
09) If there is a conflict, the one set by the user is registered with priority (110), and the conflicting past information search conditions are deleted.

The information retrieval condition does not have to be a word such as a keyword, but may be a sentence. When the keyword setting is completed, the information retrieval operation is started (11
1), the agent autonomously checks various channels of information sources (for example, Internet sites and teletext), searches for a channel including a keyword, and presents it to the user.

As described above, the information retrieving method using the agent according to the first embodiment of the present invention has an excellent effect in that the user can save the trouble of setting the condition for the information retrieval.

FIG. 2 shows a processing flow for inputting additional information search conditions by voice dialogue in the information search method using an agent according to the second embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

When the user thinks that the presented information search keyword (condition) is insufficient, the user requests additional registration of the information search keyword by voice dialogue (20
1) You can. Thereby, for example, the user can additionally register the information search condition even while driving.

It is to be noted that the keywords are checked for overlap or inconsistency with those registered in the past (205).
If the same keyword exists in the past, the registration is canceled (206) and it is checked whether the meaning conflicts (20).
7) In case of conflict, the user's setting is preferentially registered (208), and the conflicting past information search conditions are deleted.

The information retrieval condition does not have to be a word such as a keyword, but may be a sentence. After the additional registration of keywords (209), the agent is free to autonomously check various channels of information sources (for example, Internet sites and teletext) to search for a keyword-containing one, Present to the user.

As described above, the information retrieving method using the agent according to the second embodiment of the present invention has an excellent effect in that the user can save the trouble of setting the condition for the information retrieval.

FIG. 3 shows a processing flow for inputting additional information search conditions by a sentence in the information search method using the agent according to the third embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

When the user thinks that the presented information search keyword is insufficient, the user can additionally register the information search keyword. First, the user inputs a sentence relating to information he / she wants to know from a remote controller or the like (3
03). For example, when "What did Prime Minister Moribuchi say yesterday?" Is input, the meaning of the input sentence is analyzed (304), and the user understands what he or she really wants to do (3).
05), generate an information search keyword. At this time, it is checked whether or not the keyword overlaps or is inconsistent with the keyword registered in the past (309). If there is the same keyword in the past, the registration is stopped (310), and it is checked whether or not there is a semantic conflict (311). If there is a conflict, the one set by the user is preferentially registered (312), The inconsistent past information retrieval conditions are deleted.

The information retrieval condition does not have to be a word such as a keyword but may be a sentence. After the additional registration of keywords (313), the agent is free to autonomously check various channels of information sources (for example, Internet sites and teletext) to search for a keyword-containing channel. Present to the user.

As described above, the information retrieving method using the agent according to the third embodiment of the present invention has an excellent effect in that the user can save the trouble of setting the condition for information retrieval.

FIG. 4 shows a processing flow in the information retrieval method using an agent according to the fourth embodiment of the present invention, in which an agent reading the retrieved information and an agent having a dialogue are separated.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, the agent automatically appears when a user calls (401) or when a constraint condition such as that the memory capacity of the user is likely to be exceeded (403) is satisfied. If the agent is presented from the beginning, it will move actively by satisfying the above conditions. The agent autonomously checks various channels of the information source (for example, Internet sites and teletext) and searches for the one that contains the keyword, but the meaning of the search result is shown before the presentation to the user. Understand (404) and find the word that is the focus.

The meaning of the focused word is found in the language dictionary on the memory of the agent, and a sentence expressing an impression corresponding to it is generated. Since the sentence expressing this impression can be set in advance, it does not necessarily have to be text-to-speech synthesis, and may be recorded / reproduced voice. For example,
In the case of a high school girl agent, when the focus is "exit", the impression according to the meaning such as "Hey, listen to this. Terrible!" Is stated (405). After this, the agent switches to a read-only agent (406). The read-only agent reads the retrieved information using text-to-speech synthesis. Such alternation continues until no information is retrieved.

As described above, the information retrieving method using the agent according to the fourth embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 5 shows a processing flow for selecting a dialogue strategy based on the user's emotion estimation for smoothing the dialogue with the user in the information retrieval method using the agent according to the fifth embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, after one turn of dialogue with the user (5
03), the success or failure of the dialogue is recorded (504), the pattern of the dialogue result is extracted (505), and the state of the user is estimated (506) from the dialogue history.

For example, it is possible to easily estimate (506) whether the user is "joyful" or "angry" by recording (504) whether the dialogue for the past several turns has succeeded or failed. "Success" if there is a lot of success, "anger" if there is little success
Because it is. If the user is "joyful", the conversation can be carried out efficiently with a simple expression, but if the user is "angry", the expression is changed to a polite expression, and the agent's facial expression is changed to something apologetic. The utterance method and dialogue strategy are selected according to the emotion of (507).

As described above, the information retrieving method using the agent according to the fifth embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 6 is a flowchart showing the process of selecting a dialogue strategy based on the user's emotion estimation for smoothing the dialogue with the user in the information retrieval method using the agent according to the sixth embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, after one turn of dialogue with the user (6
03), the user's voice situation is analyzed (604), the estimated emotion of the user is extracted (605), and the current emotional state of the user is estimated from the voice quality of the user (606).

For example, when the average pitch frequency is slightly high and the ending time is long, the user is often happy. On the contrary, the average pitch frequency is low, the overall power is increased, and the ending time is increased. When it is short, it is often "anger". Such estimation is performed for each dialogue turn, and the estimation results are accumulated and used for emotion estimation. If the user is "joyful", the conversation can be carried out efficiently with a simple expression, but if the user is "angry", the expression is changed to a polite expression, and the agent's facial expression is changed to something apologetic. A speech method and a dialogue strategy are selected according to the emotion of (607).

As described above, the information retrieving method using the agent according to the sixth embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 7 shows a processing flow of selecting a dialogue strategy based on a user's facial expression for facilitating dialogue with the user in the information retrieval method using the agent according to the seventh embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, after one turn of dialogue with the user (7
03), the user's facial expression is analyzed (704), the estimated emotion of the user up to now is extracted (705), and the emotional state of the user is estimated from the facial expression of the user (706).

For example, when both eyes are thin and both ends of the mouth are raised, it is often "joyful", and when the distance between both eyebrows is short and the mouth is slightly open, it is "angry". In many cases. Such estimation is performed for each dialogue turn, and the estimation results are accumulated and used for emotion estimation. User is happy
In the case of, the conversation can be carried out efficiently with a simple expression, but in the case of “anger”, the expression was changed to a polite expression, the expression of the agent was changed to something apologetic, and it was adjusted to the estimated user's emotion. Select utterance method and dialogue strategy (7
07) is performed.

As described above, the information retrieving method using the agent according to the seventh embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 8 is a process of selecting a dialogue strategy based on the user's voice quality, facial expression, and dialogue history for facilitating dialogue with the user in the information retrieval method using an agent according to the eighth embodiment of the present invention. The flow is shown.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, after one turn of dialogue with the user (8
03) Successful or unsuccessful dialogue, recording the user's voice and facial expression (804), and extracting the dialogue history up to the present, voice situation, facial expression (805), from the user's voice quality, facial expression, and dialogue result. The emotional state of the user is estimated (806).

For example, if the result of the dialogue is “success”, the voice quality is “soft”, and the facial expression is “neutral”, the user is often satisfied, and conversely, the result of the dialogue is unsuccessful and the voice quality is “neutral”. "Angry" is often "anger" when the expression is "angry".
Such estimation is performed for each dialogue turn, and the estimation results are accumulated and used for emotion estimation. When the user is "joyful", the conversation can be efficiently promoted with a simple expression, but "anger"
In this case, the expression is changed to a polite expression, the expression of the agent is changed to something apologetic, and the utterance method and the dialogue strategy that match the estimated emotion of the user are selected (807).

As described above, the information retrieving method using the agent according to the eighth embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 9 shows a processing flow of autonomous information retrieval in the information retrieval method using the agent according to the ninth embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, the agent seeks information containing the information retrieval keyword that it has and automatically changes the information channels CH one after another (903). As a result, when the search of all the information channels is completed (905), the maximum number of registered information, the maximum memory size for storing information, or the search end condition of a preset time is satisfied (906), the search operation is interrupted. , The search information up to that point is transmitted to the user by synthesized voice (908). Of course, the user said, "Hey,
Please tell me ○○. It goes without saying that the search information up to that point can be transmitted to the user in a synthesized voice in response to a call such as "."

As described above, the information retrieving method using the agent according to the ninth embodiment of the present invention is also advantageous in that the user's information retrieval work can be facilitated.

FIG. 10 shows a processing flow for referring to past search results in the information search method using the agent according to the tenth embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, the agent searches for information (100
2) Then, the retrieval of the first search result (1003)
Is performed and compared with the past search result (1004), and there is information similar to this information (1005), the past search result storage unit is examined.

When similar information is found, the fact is notified to the user (1006) and the retrieval result is read aloud (100
7). For example, when the stock price news is searched, the stock searched for is read out again after notifying the similarity of the past similar information such as "The stock went down like three days ago." It is more efficient not to read past information unless read out from the user. Such processing is performed in all searches.

As described above, the information retrieving method using the agent according to the tenth embodiment of the present invention has an excellent effect in that the user's interest can be maintained.

FIG. 11 shows a processing flow for referring to past search results in the information search method using the agent according to the eleventh embodiment of the present invention.

The operation of the information retrieval method using this agent will be described with reference to FIG.

First, the agent searches for information (110
2) If so, is there any similar information to this information (110)?
6) Examine the past search result storage unit. The determination of similarity is the genre of the hit search keyword, the noun-object pair of the search condition, the superordinate concept of the search keyword, or the focus of the searched information. The hit information search keyword itself can of course be used for similarity determination, but the probability of finding similar information is increased by using these larger concepts. When similar information is found, the fact is notified to the user (1107) and the search result is read aloud (1108).

For example, when a sportsman agent is used, even if information including "tennis" in the "sports" genre is found at this time, before introducing it, the "sports" genre is changed to "baseball". If the keyword "is included", "Speaking of which, yesterday was a perfect match for Nomo," he reads the information about the tennis retrieved after reporting similar information in the past.

The verb-object pair of the information retrieval condition is
For example, when "tell me nearby ramen shop" is input as the information search condition, "ramen shop" and "teach" are extracted as a pair. If you search the accumulated information in the past and find the information that was found under the information search condition of "tell me a delicious ramen shop" last week, this is extracted because it has the same pair, and the agent "last week, delicious ramen shop" I read the information about the ramen shop found by searching at the moment. In this way, the user can recall his last week's experience at a ramen shop and review his current behavior.

The superordinate concept is different in that it has a genre and a hierarchy. For example, the generic concept of "ramen restaurant" is "Chinese restaurant", and the generic concept of "Chinese restaurant" is "place to eat". Therefore, in the example of a ramen shop above, if there was information found by searching in the "curry shop" last week, if you search for the keyword "ramen shop" at the moment, the same concept as "dining place"
I found information about a curry shop, and the agent said, "Speaking of that, I went to a curry shop last week." By doing so, the user can recall the experience at the curry shop last week and review the behavior at the present time.

The focus summarizes the retrieved information and can be extracted as the most frequently occurring word. The agent looks for past accumulated information containing this word and reports it to the user when found. For example, it is assumed that the business agent analyzes the information retrieved using “crude oil” as a keyword and extracts “price increase” as the focus.
The agent searches for information that has "price information" as the focus among the accumulated information in the past, and when he finds information on housing land price information, he said, "Speaking of that, the house price also rose last week." Then, read the information on the price increase of crude oil retrieved at the moment. By doing so, the user can gain a profit by anticipating the inflation from the rise in the house price last week and the rise in the crude oil price at the present time, and taking actions in accordance with the inflation.

By the way, it is more efficient not to read past information itself unless requested by the user, and it is sufficient to give an outline.

Such processing is performed in all searches.

As described above, the information retrieving method using the agent according to the eleventh embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 12 is a block diagram of an information retrieving apparatus using an agent according to the twelfth embodiment of the present invention.

As shown in FIG. 12, an information retrieval device using this agent includes a central processing unit 1201 and an agent display unit 1202 which displays a list of selectable agents and is generally composed of a display. An audio device including a display means 1202 and a bus interface circuit 1203, an input means 1204 generally composed of a microphone or a remote controller, an agent selection means 1205, a search keyword addition means 1206, and a speaker in general. Output means 1207, agent specific information search conditions, information search condition storage means 1208 which is generally a memory device, information search means 1209, and system bus 1210.

The operation of the information retrieving apparatus using the agent configured as described above will be described.

Information Retrieval Device Using This Agent First, the present device displays a plurality of agent images acting on behalf of the information retrieval work on behalf of the user and their characteristics on the display device 1202, and the user can display the images and the characteristic explanations. Select agents who are likely to have a tendency similar to your interests. For example, in the case of "Oyaji Agent", the image of an old man who seems to be stubborn and the characteristics such as "Slightly stubborn. I am interested in political economy. Weak to grandchildren and wives." If you want to request information such as "Father's Agent", you can expect that you can get almost the information you want.

Then, the user selects the name or number using the input means 1204. When the input means 1204 is a voice, the input means 1204 includes a voice recognition device. By the agent selecting means 1205, only the selected agent remains and the other agents are erased. The agent selecting means 1205 stores the name or number of this agent. Each agent has a unique keyword for information retrieval in the information retrieval condition storage means 1208. For example, for a business agent, "salary" and "business"
Words such as "selling" are stored in advance as information search keywords.

When the user uses the input means 1204 to present an information retrieval keyword specific to the agent, a list of keywords is displayed on the display device 1202. In some cases, it is also effective to read the registered contents from the output means 1207 by using a synthetic voice or the like. The user
When the user thinks that the presented information search keyword is insufficient, the user can additionally register the information search keyword. At that time, the central processing unit 1201 checks whether or not the keywords are the same as those registered in the past or have no contradiction.

If there is the same keyword in the past, the registration is stopped. If there is a semantic conflict, the user's setting is prioritized and the conflicting past information search conditions are deleted. The information search condition does not have to be a word such as a keyword, but may be a sentence. Once the set of keywords is complete, the agent autonomously checks various channels of the information source (eg, Internet sites, teletext, etc.) and looks for the one that contains the keyword and presents it to the user.

As described above, the information retrieving apparatus and the information retrieving apparatus using the agent according to the twelfth embodiment of the present invention have an excellent effect in that the user can save the trouble of setting the information retrieval condition.

FIG. 13 is a block diagram of an agent-specific information search condition and a newly added information search condition determination unit 1300 in the agent-based information search device according to the thirteenth embodiment of the present invention.

As shown in FIG. 13, this judgment unit 1300
The sequence control means 1301, the input means 1302, the comparison judgment means 1303, the thesaurus 1304, the agent-specific information search condition storage means 1305, and the gate means 1306 are included.

The operation of the judgment unit 1300 for the information search condition peculiar to the agent and the information search condition to be newly added in the information search apparatus using the agent thus configured will be described.

When the user thinks that the information retrieval condition (keyword or sentence) of the twelfth embodiment is insufficient, the input means 1302 is used to retrieve the information newly expressed by the information retrieval keyword or sentence. Enter the condition. This keyword or sentence is a direct information retrieval condition storage means 13
It is not sent to 05, but is temporarily stored in the gate means 1306. When the sequence control means 1301 learns that a keyword or sentence has been input by the notification from the input means 1302, it activates the comparison / determination means 1303, and compares the input with those already stored in the information search condition storage means 1305. To do.

At this time, in addition to simple comparison, both the input and the stored one are set as a set of related words in the thesaurus dictionary 1304, and it is determined whether or not the sets have the same element. As a result, if it is determined that they are not the same, an instruction is issued from the sequence control means 1301 to transfer the input accumulated in the gate means 1306 to the information search condition storage means 1305.

Further, the comparison and determination means 1303 also determines the consistency between the input and the existing information search condition. That is,
The conditions of "high price" and "low price" are meaningless when they exist at the same time. When such a contradiction is found, the comparison and determination means 1303 notifies the output means 1307 that there is a contradiction. Since the output means 1307 has a contradiction in the synthesized voice or the like, the user's input is prioritized and the existing Notifies the user that the condition has been deleted.

As described above, the new information search condition input by the user is automatically compared with the existing information search condition,
Consistency and redundancy are checked and overwritten when inconsistent. If there is no inconsistency, it will be additionally registered.

As described above, the information retrieving apparatus and the information retrieving apparatus using the agent according to the thirteenth embodiment of the present invention have an excellent effect in that the user can save the trouble of setting the information retrieval condition.

FIG. 14 is a block diagram showing a case where a voice dialogue is performed as the information retrieval condition adding means 1400 in the information retrieval apparatus using the agent according to the fourteenth embodiment of the present invention.

As shown in FIG. 14, the information retrieval condition adding means 1400 includes a sequence control means 1401, a voice recognition means 1402, a dialogue management means 1403, a comparison and determination means 1404,
A thesaurus dictionary 1405, information retrieval condition storage means 1406,
It is composed of gate means 1407.

The operation in the case where a voice dialogue is performed as the information search condition adding means 1400 of the information search device using the agent thus configured will be described.

When the user thinks that the information retrieval condition (keyword or sentence) of the twelfth embodiment is insufficient, the voice recognition means 1402 is used, and information newly expressed by the information retrieval keyword or sentence is used. Enter the search conditions.
This keyword or sentence is a direct information retrieval condition storage means.
It is not sent to 1406 but temporarily stored in the gate means 1407.

At this time, as a phenomenon peculiar to voice recognition, there are cases where erroneous recognition and rewording are necessary. Dialogue management means
1403 sees the result notification from the voice recognition means 1402, and when the likelihood of the recognition result is not sufficient, uses the voice synthesis means 1408 to request the user to re-voice. When the sequence control means 1401 learns that a valid keyword or sentence has been recognized by the notification from the dialogue management means 1403, it activates the comparison / determination means 1404, and inputs and stores it in the information retrieval condition storage means 1406. Compare things. At that time, in addition to simple comparison, both the input and the stored one are set as a set of related words in the thesaurus dictionary 1405, and it is determined whether or not the sets have the same element. As a result, if it is determined that they are not the same, the sequencer issues a command to transfer the input stored in the gate means 1407 to the information search condition storage means 1406.

Further, the comparison / judgment means 1404 judges the consistency between the input and the existing information search condition. That is,
The conditions of "high price" and "low price" are meaningless when they exist at the same time. When such a contradiction is found, the comparison and determination means 1404 notifies the voice synthesizing means 1408 that there is a contradiction. Since the voice synthesizing means 1408 has a contradiction, the user's input is prioritized and the existing condition is Notify the user that it was deleted.

As described above, the new information search condition input by the user is automatically compared with the existing information search condition,
It is registered only when there is no problem after checking consistency and redundancy.

As described above, the information retrieving apparatus and the information retrieving apparatus using the agent according to the fourteenth embodiment of the present invention have an excellent effect in that the user can save the trouble of setting the information retrieval conditions.

FIG. 15 shows a block diagram in the case where a sentence is input as the information retrieval condition adding means 1500 in the information retrieval apparatus using the agent according to the fifteenth embodiment of the present invention.

As shown in FIG. 15, the information retrieval condition adding means 1500 includes a sequence control means 1501 and an input means 15.
02, syntax analysis means 1503, meaning understanding means 1504, keyword extraction means 1505, comparison determination means 1506, information search condition storage means 1507, gate means 1508, response sentence generation means 1509, and output means 1510. It consists of and.

The operation when a sentence is input as the information retrieval condition adding means 1500 of the information retrieval apparatus using the agent thus configured will be described.

When the user thinks that the information retrieval condition (keyword or sentence) of the twelfth embodiment is insufficient, the input means 1502 is used to newly retrieve the information retrieval keyword or sentence. Enter the condition. Generally, there are many cases where input is performed using a remote control. This keyword or sentence is not directly sent to the information search condition storage means 1507, but is temporarily stored in the gate means 1508. At this time, as a phenomenon peculiar to sentence input, there may be a case where re-input is necessary due to an error or the like. From the input decomposed into units close to words by the syntactic analysis means 1503 and the meaning understanding means 1504, keywords are extracted by the keyword extraction means 1505.

At this time, the syntactic analysis means 1503 and the meaning understanding means
1504 sequentially transfers the score representing the validity of the analysis to the next keyword extracting means 1505. If the sequence control means 1501 sees the result of the keyword extraction means 1505 that received these, and if there is an error, the sequence control means 1501 uses the response sentence generation means 1509 to request the user to re-input. When the sequence control means 1501 learns that a valid keyword has been extracted by the notification from the keyword extraction means 1505, it activates the comparison / determination means 1506 to input and already enter the information retrieval condition storage means.
Compare what is stored in the 1507. At that time, in addition to simple comparison, both the input and the stored one are set as a set of related words in the thesaurus dictionary 1511 and it is determined whether or not the sets have the same element.

As a result, if it is determined that they are not the same, an instruction is issued from the sequence control means 1501 to transfer the input stored in the gate means 1508 to the information retrieval condition storage means 1507. Further, the comparison and determination means 1506 also determines the consistency between the input and the existing information search condition. In other words, the conditions "high price" and "low price" are
If they exist at the same time, it becomes a meaningless condition. When such a contradiction is found, the comparison / determination means 1506 outputs the output means 15
10 is notified that there is a contradiction, and the output means 1510 prioritizes the user's input because of the contradiction, and notifies the user that the existing condition is deleted.

In this way, the new information search condition input by the user is automatically compared with the existing information search condition,
It is registered only when there is no problem after checking consistency and redundancy.

As described above, the information retrieving apparatus and the information retrieving apparatus using the agent according to the fifteenth embodiment of the present invention have an excellent effect in that the user can save the trouble of setting the information retrieval condition.

FIG. 16 shows a block diagram when switching agents in the information search apparatus using agents according to the sixteenth embodiment of the present invention.

In FIG. 16, an information retrieval device using this agent is composed of a voice recognition means 1601 and a syntax analysis means.
1602, meaning understanding means 1603, dialogue management means 1604, information retrieval means 1605, information collection means 1606, received information storage means 1607, agent switching means 1608, recorded voice DB 1609, text voice synthesis means 1610 The image generation means 1611, the output switching means 1612, and the speaker 1613.

The operation of switching the agents in the information search apparatus using the agents thus configured will be described.

First, when there is a constraint condition such as a call from the user through the voice recognition means 1601 or the memory capacity of the agent itself is about to be exceeded,
The agent automatically appears through the image generation means 1611. If the agent is presented from the beginning, it will move actively by satisfying the above conditions. The agent autonomously checks various channels of an information source (for example, an Internet site or teletext), and the information retrieval means 1605 includes the information retrieval keyword for the information retrieval condition (keyword) that the agent itself has. The information retrieving means 1605 finds the keyword that is the focus of the retrieved information before presenting it to the user.

Based on this, the dialogue management means 1604 selects from the recorded voice database 1609 the recorded voice that expresses the impression associated with this focused keyword. The sentence expressing this impression can be decided in advance,
Text-to-speech synthesis is not necessarily required, and recording / playback voice may be used. For example, in the case where the focus is "exit", in the case of a high school girl agent, he states, "Hey, listen to this. Awful!" After this, the agent is switched to the agent specialized for reading by the agent switching means 1608. The read-only agent reads the retrieved information using text-to-speech synthesis. This change is continued until there is no information retrieved by the information retrieval means 1605.

As described above, the information retrieving apparatus using the agent according to the sixteenth embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 17 is a block diagram of an information search device using an agent according to the seventeenth embodiment of the present invention.

In FIG. 17, the information retrieval apparatus using this agent is composed of a voice recognition means 1701 and a syntax analysis means.
1702, meaning understanding means 1703, dialogue management means 1704, information retrieval means 1705, information collection means 1706, received information storage means 1707, agent switching means 1708, recorded voice database 1709, text voice synthesis means 1710,
The image generating means 1711, the output switching means 1712, the speaker 1713, and the common sense database 1714 are included.

The operation of the information search device using the agent thus configured will be described.

First, when there is a constraint condition such as a call from the user through the voice recognition means 1701 or the memory capacity of the agent itself is about to be exceeded,
The agent automatically appears through the image generation means 1711. If the agent is presented from the beginning, it will move actively by satisfying the above conditions. The agent autonomously checks various channels of an information source (for example, Internet site or teletext), and the information retrieval means 1705 includes the information retrieval keyword regarding the information retrieval condition (keyword) that the agent itself has. There is a morpheme in which the searched information is pre-processed by the text-to-speech synthesis means 1710 in order to extract the focused keyword from the searched information before presenting it to the user. Then, the focus is extracted by the syntactic analysis means 1702 and the meaning understanding means 1703.
When the meaning understanding means has a plurality of focal points, the common sense database 1714 is used to narrow the focal points to one.

Based on this, the dialogue management means 1704 selects from the recorded voice database 1709 the recorded voice that expresses the impression associated with this focused keyword. The sentence expressing this impression can be decided in advance,
Text-to-speech synthesis is not necessarily required, and recording / playback voice may be used. For example, in the case where the focus is "exit", in the case of a high school girl agent, he states, "Hey, listen to this. Awful!" After this, the agent is switched by the agent switching means 1708 to an agent specialized in reading. The read-only agent reads the retrieved information using text-to-speech synthesis. Such a change is continued until there is no information retrieved by the information retrieval means 1705.

As described above, the information retrieving apparatus using the agent according to the seventeenth embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 18 is a block diagram of an information search device using an agent according to the eighteenth embodiment of the present invention.

In FIG. 18, an information retrieval apparatus using this agent is composed of a voice recognition means 1801 and a syntax analysis means.
1802, meaning understanding means 1803, dialogue management means 1804, dialogue history database 1805, emotion estimation means 1806, information collection means 1807, received information storage means 1808, information search means 1809, agent switching means 1810 A voice recording database 1811, a text voice synthesizing means 1812, an output switching means 1813, a speaker 1814, and an image generating means.
It consists of 1815 and.

The operation of the information retrieving apparatus using the agent thus configured will be described.

First, the result of the one-turn dialogue with the user, which is formed by the output of the text-to-speech synthesis means 1812 in response to the input from the voice recognition means 1801 (or the opposite combination), is recorded in the dialogue history database 1805, and then the dialogue history is recorded. The emotion state of the user is estimated from the database 1805 using the emotion estimation means 1806. For example, it can be easily estimated whether the user is "joyful" or "angry" depending on whether the conversation in the past several turns has succeeded or failed. This is because if there is a lot of success, it means "joy," and if there is little success, it means "anger." When the user is "joyful", the dialogue management means 1804 advances the dialogue efficiently with a simple expression, but when it is "angry", the dialogue is switched to a polite expression and the image generation means 1815 makes the expression of the agent apologetic. Replace with something that looks good.

As described above, the information retrieving apparatus using the agent according to the eighteenth embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 19 is a block diagram of an information search device using an agent according to the nineteenth embodiment of the present invention.

In FIG. 19, an information retrieval apparatus using this agent is composed of a voice recognition means 1901 and a syntax analysis means.
1902, meaning understanding means 1903, dialogue management means 1904, emotion estimation means 1905, information collection means 1906, received information storage means 1907, information retrieval means 1908, agent switching means 1909, voice recording database 1910 And a text / voice synthesizing means 1911, an output switching means 1912, a speaker 1913, and an image generating means 1914.

The operation of the information search device using the agent thus configured will be described.

First, with respect to the input from the voice recognition means 1901, the emotion estimation means 1905 is used to estimate the emotional state of the user. For example, if the average pitch frequency is slightly higher and the ending time is longer, the user is often “happy”. Conversely, if the average pitch frequency is low, the overall power is increased, and the ending time is short. Often "angry". When the user is "joyful", the dialogue management means 1904 can efficiently proceed with the dialogue by using a simple expression, but when the user is "angry", the dialogue is switched to a polite expression and the image generation means 1914 makes the expression of the agent apologetic. Replace with something that looks good.

As described above, the information retrieving apparatus using the agent according to the nineteenth embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 20 is a block diagram of an information search device using an agent according to the twentieth embodiment of the present invention.

In FIG. 20, an information retrieval apparatus using this agent is a speech recognition means 2001 and a syntax analysis means.
2002, meaning understanding means 2003, dialogue management means 2004, image recognition means 2005, emotion estimation means 2006, information collection means
2007, received information storage means 2008, and information retrieval means 2009
It comprises an agent switching means 2010, a recorded voice database 2111, a text voice synthesizing means 2012, an output switching means 2013, a speaker 2014 and an image generating means 2015.

The operation of the information search apparatus using the agent thus configured will be described.

First, the emotion estimation means 2006 is used to estimate the emotional state of the user with respect to the facial expression of the user input through the image recognition means 2005 at the same time as the input from the voice recognition means 2001. For example, if both eyes are thin and both ends of the mouth are raised, it is often "joyful", and the interval between both eyebrows is short, and if the mouth is slightly open, it is often "angry". . When the user is "joyful", the dialogue management means 2004 efficiently advances the dialogue with a simple expression, but when the user is "angry", it switches to a polite expression and the image generation means 2015 makes the expression of the agent sorry. Replace with something that looks good.

As described above, the information retrieving apparatus using the agent according to the twentieth embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 21 is a block diagram of an information retrieval system using an agent according to the 21st embodiment of the present invention.

In FIG. 21, an information retrieval apparatus using this agent is composed of a voice recognition means 2101, a syntax analysis means.
2102, meaning understanding means 2103, dialogue management means 2104, image recognition means 2105, emotion estimation means 2106, and information collection means
2107, received information storage means 2108, and information retrieval means 2109
An agent switching unit 2110, a recorded voice database 2111, a text voice synthesizing unit 2112, an output switching unit 2113, a speaker 2114, and an image generating unit 2115.

The operation of the information search device using the agent thus configured will be described.

First, the emotion estimation hand 2106 is used to estimate the emotional state of the user with respect to the facial expression of the user input through the image recognition means 2105 at the same time as the input from the voice recognition means 2101. For example, if the dialogue result is “success”, the voice quality is “soft”, and the facial expression is “neutral”, the user is often satisfied, and conversely, the dialogue result is failure and the voice quality is “angry”.
When the expression is "angry", it is often "angry". When the user is "joyful", the dialogue management means 2104 efficiently advances the dialogue with a simple expression, but when the user is "angry", it switches to a polite expression and the image generation means 2115 makes the expression of the agent apologetic. Replace with something that looks good.

As described above, the information retrieving apparatus using the agent according to the twenty-first embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 22 is a block diagram of an information retrieving apparatus using an agent according to the 22nd embodiment of the present invention.

In FIG. 22, an information search apparatus using this agent is a sequence control means 2201, an input means 2202, a comparison / determination means 2203, and a thesaurus dictionary 2204.
An information retrieval condition storage means 2205, a gate means 2206,
Output means 2207, information retrieval means 2208, and time monitoring means 22
09, text-to-speech synthesis means 2210, received information storage means
2211, information collecting means 2212, and a plurality of information sources 2213.

The operation of the information search apparatus using the agent thus configured will be described.

First, the agent seeks information including the information retrieval keyword possessed by itself, replaces the plurality of information sources 2213 by the information collecting means 2212 one after another, and the received information storing means
The information stored in 2211 is searched by the information search means 2208. The time monitoring means 2209 monitors this search operation,
When the time required for the search set in advance in the sequence control means 2201 is exceeded, the search work is terminated, and the search result is transmitted to the user by the text-to-speech synthesis means 2210.

As described above, the information retrieving apparatus using the agent according to the twenty-second embodiment of the present invention also has an excellent effect in that the user's information retrieval work can be facilitated.

FIG. 23 is a block diagram of an information retrieving apparatus using an agent according to the 23rd embodiment of the present invention.

In FIG. 23, the information retrieval apparatus using this agent is a sequence control means 2301, an input means 2302, a comparison / determination means 2303, and a thesaurus dictionary 2304.
An information retrieval condition storage means 2305, a gate means 2306,
The output means 2307, the information search means 2308, the search number investigation means 2309, the text-to-speech synthesis means 2310, the received information storage means 2311, the information collection means 2312, and a plurality of information sources 2313.

The operation of the information search device using the agent thus configured will be described.

First, the agent seeks information including the information retrieval keyword possessed by itself, replaces the plurality of information sources 2313 by the information collecting means 2312 one after another, and the received information storing means
The information stored in 2311 is searched by the information search means 2308. This search work is monitored by the search number investigation means 2309, and when the number of cases exceeds the upper limit number set in advance in the sequence control means 2301, the search work is terminated, and the search result is sent to the user by the text-to-speech synthesis means 2310. Tell.

As described above, the information retrieving apparatus using the agent according to the twenty-third embodiment of the present invention also has an excellent effect in that the user's information retrieval work can be facilitated.

FIG. 24 is a block diagram of an information retrieving apparatus using an agent according to the 24th embodiment of the present invention.

In FIG. 24, the information retrieval apparatus using this agent is a sequence control means 2401, an input means 2402, a comparison and determination means 2403, and a thesaurus dictionary 2404.
An information retrieval condition storage means 2405, a gate means 2406,
Output means 2407, information retrieval means 2408, memory size investigation means 2409, text-to-speech synthesis means 2410, received information storage means 2411, information collection means 2412, and a plurality of information sources 24
It is composed of 13 and.

The operation of the information search device using the agent thus configured will be described.

First, the agent seeks information including the information retrieval keyword possessed by itself, and successively replaces the plurality of information sources 2413 by the information collecting means 2412, and the received information storing means
The information stored in 2411 is searched by the information search means 2408. This search work is monitored by the memory size investigation means 2409, and when the data accumulated in the reception information storage means 2411 exceeds the upper limit size preset in the sequence control means 2401, the search work is terminated, The text-to-speech synthesizer 2410 informs the user of the search result at that time.

As described above, the information retrieving apparatus using the agent according to the twenty-fourth embodiment of the present invention also has an excellent effect in that the user's information retrieval work can be facilitated.

FIG. 25 is a block diagram of an information search device using an agent according to the twenty-fifth embodiment of the present invention.

In FIG. 25, the information retrieval apparatus using this agent is a sequence control means 2501, an input means 2502, a comparison judgment means 2503, and a thesaurus dictionary 2504.
An information retrieval condition storage means 2505, a gate means 2506,
Output means 2507, information search means 2508, search upper limit condition investigation means 2509, text voice synthesis means 2510, received information storage means 2511, information collection means 2512, and a plurality of information sources 25
13, a storage area 2514 for past received information, and a dialogue voice database 2515 of the agent.

The operation of the information search device using the agent thus configured will be described.

First, the agent is the information collecting means 2512.
Using, you can autonomously search for information that matches the information search conditions of the agent itself from multiple information sources 2513.
The storage area 2514 of the received information in the past is checked to see if there is information similar to this information. When similar information is found, the optimum voice is selected from the agent's dialogue voice database 2515, and the fact is reported to the user through the output means 2507.
The text-to-speech synthesizer 2510 is used to read the search result. For example, when searching stock price news, the stock information is reported to be similar to the stock price news like 3 days ago. Then, the searched information is read aloud. Such an operation is continued until the search upper limit condition checking means 2509 orders stop.

As described above, the information retrieving apparatus using the agent according to the twenty-fifth embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 26 is a block diagram of an information retrieving apparatus using an agent according to the 26th embodiment of the present invention.

In FIG. 26, the information retrieval apparatus using this agent is a sequence control means 2601, an input means 2602, a comparison judgment means 2603, and a thesaurus dictionary 2604.
An information retrieval condition storage means 2605, a gate means 2606,
Output means 2607, information search means 2608, search upper limit condition investigation means 2609, text voice synthesis means 2610, received information storage means 2611, information collection means 2612, and a plurality of information sources 26
13 and keyword genre storage area 26 for past received information
14, a keyword genre assigning means 2615, and an agent's dialogue voice database 2616.

The operation of the information search apparatus using the agent thus configured will be described.

First, the agent is the information collecting means 2612.
Using, you can autonomously search for information that matches the information search conditions of the agent itself from multiple information sources 2613.
In order to check whether there is information similar to this information, the genre of the search keyword of the searched information is determined by the keyword genre assigning means 2615, and the keyword genre storage area 2614 of the past received information is checked. Comparison judgment means 26
If it is judged as similar in 03, an optimum voice is selected from the agent's speech voice database 2616, and the output means 26
The fact is reported to the user through 07 and the search result is read aloud using the text-to-speech synthesis means 2610. For example, when searching stock price news, the stock information is reported to be similar to the stock price news like 3 days ago. Then, the searched information is read aloud. Such an operation is continued until the search upper limit condition checking means 2609 orders the stop.

As described above, the information retrieving apparatus using the agent according to the twenty-sixth embodiment of the present invention also has an excellent effect in that the interest of the user can be maintained.

FIG. 27 is a block diagram of an information retrieving apparatus using an agent according to the 27th embodiment of the present invention.

In FIG. 27, an information retrieval apparatus using this agent is a sequence control means 2701, an input means 2702, a comparison judgment means 2703, and a thesaurus dictionary 2704.
An information retrieval condition storage means 2705, a gate means 2706,
Output means 2707, information search means 2708, search upper limit condition investigation means 2709, text voice synthesis means 2710, received information storage means 2711, information collection means 2712, and a plurality of information sources 27
13, a storage area 2714 for verbs and objects of information retrieval conditions received in the past, and a means for extracting verbs and objects of information retrieval conditions
2715 and an agent's dialogue speech database 2716.

The operation of the information search device using the agent thus configured will be described.

First, the agent is the information collecting means 2712.
Using, you can autonomously search for information that matches the information search conditions of the agent itself from multiple information sources 2713.
The extraction means 2715 extracts the verb and the object of the information retrieval condition retrieved to find out whether there is information similar to this information, and examines the storage area 2714 of the verb and the object of the information retrieval condition received in the past. If the comparison and determination means 2703 determines that the two are similar, the optimum speech is selected from the agent speech database 2716, the fact is reported to the user through the output means 2707, and the text-to-speech synthesis means 2710 is selected.
Read out the search results using. For example, when searching stock price news, the stock information is reported to be similar to the stock price news like 3 days ago. Then, the searched information is read aloud. Such an operation is continued until the search upper limit condition checking means 2709 commands the stop.

As described above, the information retrieving apparatus using the agent according to the twenty-seventh embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 28 is a block diagram of an information retrieval system using an agent according to the 28th embodiment of the present invention.

In FIG. 28, the information retrieval apparatus using this agent is a sequence control means 2801, an input means 2802, a comparison and determination means 2803, and a thesaurus dictionary 2804.
An information retrieval condition storage means 2805, a gate means 2806,
Output means 2807, information retrieval means 2808, search upper limit condition examination means 2809, text-to-speech synthesis means 2810, received information storage means 2811, information collection means 2812, and a plurality of information sources 28
13, a storage area 2814 for a higher-level concept of a search keyword for past received information, and a means for adding a higher-level concept of a search keyword
2815 and the agent's dialogue voice database 2816.

The operation of the information search apparatus using the agent thus configured will be described.

First, the agent is the information collecting means 2812.
Using, you can autonomously search for information that matches the information search conditions of the agent itself from multiple information sources 2813.
In order to find out whether there is information similar to this information, the superordinate concept of the search keyword of the searched information is given as the adding means 2815.
Then, the superordinate concept storage area 2814 of the search keyword of the past received information is searched.

When the comparison / judgment means 2803 determines that the speech is similar, the optimum voice is selected from the speech voice database 2816 of the agent, the fact is reported to the user through the output means 2807, and the text voice synthesis means 2810 is used to retrieve the search result. Read aloud. The superordinate concept is different in that it has a genre and a hierarchy. For example, the generic concept of "ramen restaurant" is "Chinese restaurant", and the generic concept of "Chinese restaurant" is "place to eat".

Therefore, in the example of the ramen shop above,
If there is information found by searching in the "curry shop" last week, if you search for the keyword "ramen shop" at the present time, information on the curry shop that is the same "dining place" is found, and the agent says " Speaking of which,
Last week you went to a curry shop. ", And then read out the search results of the ramen shop at the moment. By doing so, the user can recall the experience at the curry shop last week and review the behavior at the present time. The information retrieval apparatus performs such an operation as the retrieval upper limit condition investigation means 2809.
Will continue until you command them to stop.

As described above, the information retrieving apparatus using the agent according to the twenty-eighth embodiment of the present invention also has an excellent effect in that the user's interest can be maintained.

FIG. 29 is a block diagram of an information retrieving apparatus using an agent according to the 29th embodiment of the present invention.

In FIG. 29, the information retrieval apparatus using this agent is a sequence control means 2901, an input means 2902, a comparison judgment means 2903, and a thesaurus dictionary 2904.
An information retrieval condition storage means 2905, a gate means 2906,
Output means 2907, information search means 2908, search upper limit condition investigation means 2909, text voice synthesis means 2910, received information storage means 2911, information collection means 2912, and a plurality of information sources 29
13, a past reception information focus storage area 2914, a search information focus extraction means 2915, and an agent's dialogue voice database 2916.

The operation of the information retrieving apparatus using the agent thus configured will be described.

First, the agent is the information collecting means 2912.
Using, you can autonomously search for information that matches the information search conditions of the agent itself from multiple information sources 2913.
The focus of the retrieved information is estimated by the estimation means 2915 to check whether there is information similar to this information, and the focus storage area 2914 of the past received information is examined. Comparison judgment means
If it is judged as similar in 2903, the optimum voice is selected from the agent's speech voice database 2916 and output means
The fact is reported to the user through 2907, and the search result is read aloud using the text-to-speech synthesis means 2910.

For example, it is assumed that the businessman agent analyzes the information retrieved using “crude oil” as a keyword and extracts “price increase” as the focus. The agent searched for information that had "price information" as the focus among the accumulated information in the past, and, for example, when he found information on housing land price information, he said, "By the way, the house price also rose last week." Then, read out the information on the price increase of crude oil retrieved at the present time.

[0257] By doing so, the user can be profitable by anticipating the inflation from the rise in the house price last week and the rise in the crude oil price at the present time, and taking action according to this. The information retrieval apparatus continues such an operation until the retrieval upper limit condition examination means 2909 orders the stop.

As described above, the information retrieving apparatus using the agent according to the twenty-ninth embodiment of the present invention has an excellent effect in that the user's interest can be maintained.

[0259]

INDUSTRIAL APPLICABILITY As described above, according to the present invention, the user only needs to select an agent that is considered to be close to his or her own information search condition from a plurality of agents that have preset information search conditions. The information can be obtained, and the user can obtain the information without time and effort.

[Brief description of drawings]

FIG. 1 is a flowchart of an information search method using an agent according to a first embodiment of the present invention.

FIG. 2 is a process flow diagram of inputting additional information search conditions by voice dialogue in an information search method using an agent according to the second embodiment of the present invention.

FIG. 3 is a process flow diagram of inputting additional information search conditions by a sentence in an information search method using an agent according to a third embodiment of the present invention.

FIG. 4 is a process flow chart in the information search method using an agent according to a fourth embodiment of the present invention, in which an agent that reads the searched information and an agent that interacts are separated.

FIG. 5 is a process flow chart of selecting a dialogue strategy based on estimation of a user's emotion for facilitating a dialogue with a user in an information retrieval method using an agent according to a fifth embodiment of the present invention.

FIG. 6 is a process flow chart of selecting a dialogue strategy based on estimation of a user's emotion for facilitating a dialogue with a user in an information search method using an agent according to a sixth embodiment of the present invention.

FIG. 7 is a process flow chart of selecting a dialogue strategy based on a user's facial expression for facilitating dialogue with the user in the information search method using the agent according to the seventh embodiment of the present invention.

FIG. 8 is a process flow chart of selecting a dialogue strategy based on the user's voice quality, facial expression, and dialogue history for facilitating dialogue with the user in the information search method using the agent according to the eighth embodiment of the present invention.

FIG. 9 is a processing flow chart of autonomous information search in an information search method using an agent according to a ninth embodiment of the present invention.

FIG. 10 is a processing flow chart for referring to past search results in the information search method using the agent according to the tenth embodiment of the present invention.

FIG. 11 is a processing flow chart for referring to past search results in the information search method using the agent according to the eleventh embodiment of the present invention.

FIG. 12 is a block diagram of an information search device using an agent according to a twelfth embodiment of the present invention.

FIG. 13 is a block diagram of a judgment unit for determining information search conditions unique to the agent and information search conditions to be newly added in the information search device using the agent according to the thirteenth embodiment of the present invention.

FIG. 14 is a block diagram of a case where a voice dialogue is performed as an information search condition adding means in an information search device using an agent according to a 14th embodiment of the present invention.

FIG. 15 is a block diagram when a sentence is input as an information search condition adding means in an information search device using an agent according to a fifteenth embodiment of the present invention.

FIG. 16 is a block diagram of an agent switching unit in an information search device using an agent according to a sixteenth embodiment of the present invention.

FIG. 17 is a block diagram of an information search device using an agent according to a seventeenth embodiment of the present invention.

FIG. 18 is a block diagram of an information search device using an agent according to an eighteenth embodiment of the present invention.

FIG. 19 is a block diagram of an information search device using an agent according to a nineteenth embodiment of the present invention.

FIG. 20 is a block diagram of an information search device using an agent according to a twentieth embodiment of the present invention.

FIG. 21 is a block diagram of an information search device using an agent according to a twenty first embodiment of the present invention.

FIG. 22 is a block diagram of an information search device using an agent according to a twenty second embodiment of the present invention.

FIG. 23 is a block diagram of an information search device using an agent according to a twenty-third embodiment of the present invention.

FIG. 24 is a block diagram of an information search device using an agent according to a twenty fourth embodiment of the present invention.

FIG. 25 is a block diagram of an information search device using an agent according to a twenty-fifth embodiment of the present invention.

FIG. 26 is a block diagram of an information search device using an agent according to a twenty sixth embodiment of the present invention.

FIG. 27 is a block diagram of an information search device using an agent according to a twenty-seventh embodiment of the present invention.

FIG. 28 is a block diagram of an information search device using an agent according to a twenty-eighth embodiment of the present invention.

FIG. 29 is a block diagram of an information search device using an agent according to a twenty ninth embodiment of the present invention.

FIG. 30 is a block diagram of a conventional information search device using an agent.

[Explanation of symbols]

1201 Central processing unit 1202 Agent display means 1203 Interface circuit, 1204, 1302, 1502, 2202, 2302, 2402, 2502, 2602, 27
02, 2802, 2902 Input means 1205 Agent selection means 1206 Search keyword addition means 1207, 1510, 2207, 2307, 2407, 2507, 2607, 2707, 28
07, 2907 Output means 1208, 1305, 1406, 1507, 2205, 2305, 2405, 2505, 26
05, 2705, 2805, 2905 Information retrieval condition storage means 1209, 1605, 1705, 1809, 1908, 2009, 2109, 2208, 23
08, 2408, 2508, 2608, 2708, 2808, 2908 Information retrieval means 1301, 1401, 1501, 2201, 2301, 2401, 2501, 2601, 27
01, 2801, 2901 Sequence control means 1303, 1404, 1506, 2203, 2303, 2403, 2503, 2603, 27
03, 2803, 2903 Comparison judgment means 1304, 1405, 2204, 2304, 2504, 2604, 2704, 2804, 29
04 Thesaurus dictionary 1402, 1601, 1701, 1801, 1901, 2001, 2101 Speech recognition means 1403, 1604, 1704, 1804, 1904, 2004, 2104 Dialog management means 1503, 1602, 1702, 1802, 1902, 2002, 2102 Syntax analysis Means 1504, 1603, 1703, 1803, 1903, 2003, 2103 Meaning understanding means 1505 Keyword extraction means 1509 Response sentence generation means 1606, 1706, 1807, 1906, 2007, 2107, 2212, 2312, 24
12, 2512, 2612, 2712, 2812, 2811, 2912 Information collecting means 1607, 1707, 1808, 1907, 2008, 2108, 2211, 2311, 24
11, 2511, 2611, 2711, 2811, 2911 Reception information storage means 1608, 1708, 1810, 1909, 2010, 2110 Agent switching means 1609 Recording voice DB 1610, 1710, 1911, 2210, 2310, 2410, 2510, 2610, 27
10, 2810, 2910 Text-to-speech synthesis means 1611, 1711, 1815, 1914, 2015, 2115 Image generation means 1709, 1811, 1910, 2011, 2111 Recording voice database 1714 Common sense database 1805 Dialog history database 1806, 1905, 2006, 2106 Emotion Estimating means 1812, 1911, 2012, 2112, 2210, 2310 Text-to-speech synthesis means 2005, 2105 Image recognition means 2209 Time monitoring means 2313, 2213, 2413, 2513, 2613, 2713, 2813, 2913 Multiple sources 2309, 2509, 2609, 2709, 2809, 2909 Search upper limit condition checking means 2514 Past reception information storage area 2515, 2616, 2716, 2816, 2916 Agent speech database 2614 Past reception information keyword genre storage area 2615 Keyword genre assigning means 2714 Storage area for verbs and objects of search conditions of past received information 2715 Extraction means of verbs and objects of search conditions 2814 Applying means preamble of keyword 2914 focus extractor focal storage area 2915 retrieval information previously received information

Claims (40)

[Claims] 1. Presenting a plurality of predetermined agents to which unique information retrieval conditions are set individually, the agent is selected from the plurality of agents according to the information retrieval conditions desired by the user, and the agent desires An information retrieval method using an agent characterized by performing information retrieval. 2. The specific information search condition is presented sequentially by voice or in a list format on a screen according to a voice from a user or an instruction from a remote controller. Information retrieval method using agents. 3. When additionally registering the information search condition desired by the user, the information search condition desired by the user is incorporated into the existing information search condition, and added to the existing information search condition. The information retrieving method using an agent according to claim 1, wherein the additional registration is not performed when the same one as described above exists. 4. It is checked whether the information search condition additionally registered by the user is semantically inconsistent with the existing information search condition, and if the information is inconsistent, the registration is rejected, or which one is to be registered. 4. The method according to claim 3, wherein the determination is made.
Information retrieval method using the agent described in. 5. The information retrieval method using an agent according to claim 2, wherein a voice dialogue between the user and the agent is used when additionally registering an information retrieval condition desired by the user. 6. When additionally registering the information search condition desired by the user, the information search condition desired by the user is input in a natural sentence, the sentence is analyzed to extract the condition necessary for the information search, The information search method using an agent according to claim 1, wherein the information search condition is added to the table of the unique information search conditions. 7. The agent according to claim 1, wherein the agent that interacts with the user and performs a search according to an information search condition is different from an agent that reads information obtained by the information search. Information retrieval method using. 8. When the sentence of information obtained by the information search matches a sentence existing in the agent in advance, an impression corresponding to the matched content is expressed to the user. An information retrieval method using the agent described in 1. 9. The utterance content and utterance strategy of the dialogue with the user are determined according to the current emotional state of the user estimated from the history of the most recent dialogue with the user. Information retrieval method using the agent described in. 10. The utterance content and utterance strategy at the time of the dialogue with the user are determined according to the emotional state of the user estimated from the state of the user's voice. Information Retrieval Method Using Other Agents. 11. The agent according to claim 1, wherein the utterance content and the utterance strategy in the dialog with the user are determined according to the emotional state of the user estimated from the facial expression of the user. Information retrieval method using. 12. The utterance content and utterance strategy in the dialog with the user are estimated according to the emotional state of the user comprehensively estimated from the history of the dialog with the user, the state of the user's voice, and the facial expression. The information search method using an agent according to claim 1, wherein the information is determined. 13. The information search condition of the agent is used to constantly search a plurality of predetermined information sources, and the search result is reported to the user at regular time intervals. Information retrieval method using agents. 14. The information search condition of the agent is always used to search a plurality of predetermined information sources, and when the search result reaches a preset number, the user is notified of the search result. An information retrieval method using an agent according to claim 1, characterized in that. 15. The user is always searched for a plurality of predetermined information sources using the information search condition of the agent, and when the total amount of data obtained as a result of the search reaches a preset limit capacity, the user. The information retrieval method using an agent according to claim 3, wherein the result of the retrieval is reported to. 16. When storing the information obtained by the information retrieval in a storage area, the similarity with the already stored past information is judged, and when the similar information exists, the presence of the similar information is judged. The information retrieval method using an agent according to claim 1, wherein the information is reported to a user. 17. When the information obtained by the information retrieval is stored in a storage area, the obtained information and the past information are analyzed and abstracted as a set of mutually associated keywords. The information retrieval method using an agent according to claim 1, wherein the information is generated and used for similarity determination, and further, the past information is deleted after a predetermined time has elapsed. 18. When the information obtained by the information retrieval is stored in a storage area, the obtained information and the past information are analyzed and a set of superordinate concept words of the extracted keyword is generated and similar. The information search method using an agent according to claim 1, wherein the information is used for sex determination, and further, the past information is deleted after a predetermined time has elapsed. 19. When storing information obtained by the information search in a storage area, a set of keywords of both the new search result and the accumulated past search result is examined to find a verb-object pair. The information retrieval method using an agent according to claim 1, wherein the determination criterion of similarity is determined by determining whether or not a pair of both the verb and the object match each other. 20. When new information is obtained by the information search, the newly searched information is analyzed, the theme that is the focus of the information is extracted, and the similarity judgment with the accumulated past information is performed. The information retrieval method using an agent according to claim 1, wherein the information retrieval method is used for. 21. A means for holding a plurality of predetermined agents, a means for storing information retrieval conditions unique to each of the plurality of agents, a means for displaying the agent on a screen, and the displayed agent. An information retrieval apparatus using an agent, characterized by comprising: 22. A means for interpreting an instruction from the user input via a microphone or a remote control receiver,
Means for reading information retrieval conditions peculiar to the agent by recorded voice or synthesized voice according to the result of the interpretation, means for displaying a list of information retrieval conditions peculiar to the agent on the screen, the reading means, and the list display 22. The information retrieving apparatus using an agent according to claim 21, further comprising means for selectively executing the means for performing or both of them. 23. The means for extracting the information search condition from an input from a microphone or a remote control receiver and the information search condition stored in the storage means and the input information search condition are not the same. Means for determining whether the same thing exists, means for generating a sentence that reports that the same thing has been removed when the same thing exists, and means for outputting the sentence in synthetic speech ,
The information retrieval apparatus using an agent according to claim 21, further comprising: a unit that displays the sentence on a screen. 24. Meaning understanding means for understanding the meaning of an information search condition to be additionally registered by the user, meaning of the information search condition already stored in the storage means in the agent, and newly added registration Means for determining a contradiction with the meaning of the input information search condition; and means for rejecting additional registration from the user and reporting the refusal of the additional registration to the user when the contradiction is detected, or the additional registration 22. A means for inquiring of the user which one of the information search condition to be selected and the stored information search condition to be selected, and re-registering the information search condition in a storage area of the information search condition. Information Retrieval Device Using Other Agents. 25. Means for understanding the meaning of the utterance of the user by using a voice recognition means and a language analysis means including a language dictionary, and whether the utterance of the user includes an information retrieval condition desired by the user. The means for determining, a means for generating a dialogue strategy based on slot filling for pulling out information search conditions from the user, and a means for generating a synthetic voice according to the dialogue strategy. An information retrieval apparatus using the agent described in 1. 26. A means for inputting a sentence including an information search condition desired by the user, a means for analyzing the sentence by comparing it with a typical sentence template, and a means for not matching the typical sentence template. 22. The apparatus according to claim 21, further comprising: a means for analyzing an input using a language dictionary and a language analysis algorithm; and a means for extracting an information retrieval condition desired by the user from a result of the analysis. Information Retrieval Device Using Other Agents. 27. A means for generating a strategy for selectively using the agent, which interacts with the user to perform a search according to an existing information search condition, a strategy for selectively using the agent that reads out information by the search, and causes the agent to operate according to the strategy. An information retrieval apparatus using an agent according to claim 21, further comprising: means. 28. Means for performing linguistic processing of the searched information, means for understanding the meaning from the result of the linguistic processing, and means for expressing the theme of the information searched from the meaning as a network of keywords. , Means for storing common sense peculiar to the agent in the keyword network, means for determining mutual intimacy of the keyword network, and means for generating an impression sentence corresponding to the result of the determination, The information retrieval device using an agent according to claim 21, further comprising: a unit that synthesizes the impression sentence. 29. Means for monitoring a dialogue with the user and recording a dialogue history, means for estimating a current emotion of the user from the dialogue history, and a dialogue strategy for leading the emotion of the user to pleasure. 22. The apparatus according to claim 21, further comprising: a means for generating, an utterance is assembled according to the strategy, an emotional expression of a synthetic voice and a means for changing a facial expression of the character when the character is displayed on the screen. Information Retrieval Device Using Other Agents. 30. A means for analyzing the prosody of the user's voice, a means for estimating the current emotion of the user from the prosody and the utterance content of the voice, and a dialogue strategy for leading the emotion of the user to pleasure. 22. The apparatus according to claim 21, further comprising: a means for generating the speech, an utterance being assembled according to the dialogue strategy, a means for changing the emotional expression of the synthetic voice, and a means for changing the facial expression of the character when the character is displayed on the screen. Information retrieval device using the described agent. 31. A means for recognizing the facial expression of the user,
Means for estimating the emotional state of the user from the facial expression, means for generating a dialogue strategy for leading the emotion of the user to joy,
22. The agent according to claim 21, further comprising: an utterance is assembled according to the dialogue strategy, and an emotional expression of a synthetic voice and a means for changing a facial expression of the character when the character is displayed on the screen. The information retrieval device that was used. 32. Means for monitoring the dialogue with the user and recording the dialogue history, means for analyzing the prosody of the voice of the user, means for recognizing the facial expression of the user, the dialogue history, and the prosody. A means for comprehensively estimating the current emotion of the user from the content of the utterance and a combination of facial expressions of the user; a means for generating a dialogue strategy for leading the emotion of the user to happiness; 22. The information retrieval device using an agent according to claim 21, further comprising: an utterance assembling process, an emotional expression of a synthetic voice, and means for changing a facial expression of the character when the character is displayed on the screen. 33. A means for performing a search while switching a plurality of information sources according to an information retrieval condition inherent in the agent, a means for storing information as a result of the retrieval, and whether or not a predetermined time has elapsed. 22. An information search apparatus using an agent according to claim 21, further comprising: means for reporting the result of the search each time a fixed time arrives. 34. A means for performing a search while switching a plurality of information sources according to the information retrieval conditions inherent in the agent, a means for storing information on the result of the retrieval, and the number of stored cases reaches a preset number of cases. 22. The information retrieving apparatus using an agent according to claim 21, further comprising means for reporting a result of the retrieval. 35. Means for searching while switching a plurality of information sources according to information retrieval conditions inherent in the agent, means for storing information on the result of the retrieval, and means for calculating the stored total amount of data, The information search device using an agent according to claim 21, further comprising means for reporting the result of the search when the total amount of data reaches a preset upper limit value. 36. A means for storing information obtained by information retrieval, a means for determining similarity between information of past retrieval results stored in the storage means, and information for determining the similarity by the user. 22. The information retrieving apparatus using an agent according to claim 21, further comprising means for reporting to the agent. 37. Means for storing information obtained by information retrieval, means for analyzing the information obtained by the retrieval and abstracting as a set of mutually related keywords, said set of keywords and original text And storage means for storing both, a means for causing an interrupt every predetermined time, and a means for deleting the original text of the information that has been reported at the time of the interrupt. An information search device using the agent according to Item 21. 38. Means for storing information obtained by information retrieval, means for analyzing the information obtained by the retrieval and abstracting as a set of mutually associated keywords, and a superordinate concept of the keywords. A determination unit, a storage unit that stores both the superordinate concept and the original text of the search result, a unit that causes an interrupt every predetermined time, and an original text of the information that has been reported at the time of the interrupt. The information retrieval device using an agent according to claim 21, further comprising: a deleting unit. 39. When information is obtained by information retrieval,
In order to judge the similarity to the information of the accumulated past information retrieval results, the means for storing the verb and the object as a pair in the set of both keywords, and whether the pair of the verb and the object match. The information search apparatus using an agent according to claim 21, further comprising: a determination unit that determines whether or not it is. 40. When information is obtained by information retrieval,
Means for analyzing the search results obtained by the information search to determine the similarity to the accumulated information of the past search results, and means for extracting the theme of the searched information using the results of the analysis 22. Information using an agent according to claim 21, further comprising: a means for storing the theme together with an original sentence of information that is a search result; and a means for determining whether or not the theme matches. Search device.