Hitachi, Ltd.

Tadashi Hirose

Kazuo Nakao

Hiroshi Yajima

Chizuko Yasunobu

Hiroyuki Okuda

Kousuke Sakoda

Maki Tamano

Tetsuya Maruoka

Mayumi Kida

Kouichiro Ino

Yoshiya Miyagawa

Noritugu Kagaya

Keiko Shinada

Hiroshi Yoshiura

Hiroshi Tsuji

Satoshi Wakayama

Yoji Taniguchi

Tomihiko Kojima

Toyozo Aoyama

Hidehiko Akita

Tsutomu Miyairi

Youichi Takeuchi

Shuzo Tomohiro

Yoichi Takeuchi

For retrieving attribute information through image information which is not linked with the attribute information: displaying on a display screen the first image information corresponding with attribute information and second image information that has a common attribute with the first image information; and storing a correspondence between the first image information and the second image information in response to an input of an instruction to confirm the correspondence between the first image information and the second image information, and retrieving and displaying the attribute information on the basis of the stored correspondence between the first image information and the second image information as well as correspondence information between the first image information and the attribute information, in response to an input for object choice indicated by the second image information.

Retrieval method using image information

Rules registered in the electronic secretary system are defined by separating each of them into a trigger descriptor part, a condition part for object data retrieval, and an action part, and a flag indicating whether or not to fire a rule and a flag indicating whether or not to suppress firing of a rule after it is fired once are installed in each rule, and a flag for selecting whether to automatically activate the action when object data is retrieved or to activate the action after the user confirms it is installed, and an area for storing the period and timing for judging success or failure of the condition when necessary is referred to so as to change the corresponding flag to ON or OFF or to change the period and, whether to automatically execute the electronic secretary system or to execute it after a human confirms it, can be selected freely.

Rule-based electronic agent system and method thereof

A method of supporting decision-making for predicting future time-series data by using measured values of time-series data stored in a storage and knowledge stored in a knowledge base. The knowledge is related to interpretation and prediction of the time-series data. A plurality of time-series data stored in the storage device is prepared in an order of time-serial points corresponding to the time-series data. Knowledge which includes reference to the time-series data using a relative time interval from the time-series data is stored to the knowledge base. One the time-serial points of the time-series data is designated. Thereafter, the relative time interval in the knowledge is subtracting from the designated time-serial point to determine the time-serial point corresponding to the time-series data, and the future time-series data is inferred by using the knowledge and the time-series data.

Method of supporting decision-making for predicting future time-series

In a method of retrieving optimum case information for a current problem, case candidates having features of the current problem and common features are retrieved from information representing a plurality of cases stored in a case base in response to presentation of the current problem. A storing unit stores the case base, and respective cases include case problems and solutions or solving methods for the case problems. An optimum case candidate among case candidates is determined from a group of features of the current problem, a group of common features and a group of features of problems of respective case candidates in accordance with influence relation information indicating a feature group influenced by other feature groups. Case information corresponding to the optimum case candidate is retrieved from information representing a plurality of cases.

Method for retrieving case for problem and inferring solution of the

A method of supporting the decision-making on the basis of a table type data is disclosed, which comprises the processes of defining a knowhow relating to the interpretation of the table type data, storing the knowhow in a storage unit, effecting the reasoning by application of the knowhow thus defined, and evaluating the knowhow thus defined. (a) Information is applied by way of an input unit for starting any of the process of defining a knowhow, effecting the reasoning by application of the knowhow and ending the system process, (b) the information thus inputted is discriminated to execute a corresponding process of operation, and (c) if the corresponding process of operation is other than ending, the process is returned to step (a).

Method for determining a supplemental transaction changing a decided

A sequence of calling clauses and all clauses in a procedure that includes the selected clauses are displayed simultaneously, and the present point of execution is displayed being overlapped on the above displays. In particular, the program source list of a procedure being processed is displayed, and the latest execution statuses of the executed terms are graphically displayed around the terms on the displayed program source list.

Method of displaying execution trace in a logic programming language

In order to improve the efficiency of test and debug procedure based on delarative program understanding of a program described by a logical language, the operation which has been executed for each procedure is grasped. On the basis of information relating to the grasped operation, the operation range of each procedure is expressed clearly by making thick only sides which have actually operated among four sides of a box-shaped figure surrounding each head term and end body term of each clause included in each procedure.

Method for displaying operation range of logical language

In order to highly speed up the pattern matching of tree structured data in a logic programming language, the priority order is set when the data owned by the individual nodes of a tree structure are to be transversely sought, and the tree structured data are expressed in a vector type, in which they are arranged in that priority order, so that they are compared consecutively from the head for each element of the vector.

Pattern matching method for tree structured data

A computer system comprises a memory for storing communication history information between a terminal and an execution program, and an output message file for storing sets of communication history conditions and output messages corresponding to those conditions. When the terminal issues a message output request, only the output message that meets the communication history condition is selected from the output message file and outputted to the terminal.

Growing community of inventors

Komae, Japan

Tadashi Hirose