Kabushiki Kaisha Toshiba

Ryohei Orihara

Tomoko Murakami

Shigeaki Sakurai

Akihiro Suyama

Masayuki Okamoto

Hideki Kobayashi

Takashi Koiso

Ken Ueno

Naoki Imasaki

Miwako Doi

Daisuke Yamamoto

Kazushige Ouchi

Kouta Nakata

Kouichirou Mori

Yoshiaki Mizuoka

Yuichi Miyamura

Ayana Yamamoto

Youichi Kitahara

Ichiro Toyoshima

Takashi Takayanagi

Hideyuki Inoue

Kazue Nagao

According to one embodiment, a document processing apparatus includes an acquisition unit, an extractor, and a detector. The acquisition unit acquires a first document and a second document different from the first document. The extractor extracts a first structure representing a word dependence of structural elements in the first document and extracts a second structure representing a word dependence of structural elements in the second document. The detector detects a difference for each structural element from the first structure and the second structure.

Document processing apparatus, method, and program

According to an embodiment, a material recommendation apparatus includes an extractor, a creator, a matcher and a recommender. The extractor extracts information about a substance as a candidate for a material for an industrial product, a property of the substance, and a report time of the property from an electronic document. The creator creates, for each substance, time series data in which the property is associated with the report time. The matcher matches the time series data with a pattern. The recommender recommends, as the material, a substance corresponding to time series data that matches the pattern.

Material recommendation apparatus

According to one embodiment, a data connection device includes a first log storage, a second log storage, and a hardware processor. The first log storage is configured to store a plurality of first logs indicating a first behavior of a first user. The second log storage is configured to store a plurality of second logs which are independent of the plurality of first logs and indicate a second behavior of a second user. The hardware processor is configured to acquire a set of first logs indicating a specific first behavior, from the plurality of first logs, acquire a set of second logs having a feature in common with the set of first logs, and connect the set of first logs with the set of second logs.

Data connection device, method and storage medium

A device for speech input includes a speech input unit configured to convert a speech of a user to a speech signal; an angle detection unit configured to detect an angle of the speech input unit; a distance detection unit configured to detect a distance between the speech input unit and the user; and an input switch unit configured to control on and off of the speech input unit based on the angle and the distance.

Speech input device, speech recognition system and speech recognition method

A viewing behavior learning apparatus includes: a viewing history acquiring unit that acquires a viewing record that indicates an attribute of a program and start and end times during which a viewer viewed the program; a viewing history dividing unit that divides the viewing record every discretized time points that is discretized by a unit time; a viewing history replicating unit that discretizes a viewing time period denoted by the divided viewing records by interval shorter than the unit time to obtain discretized viewing time periods; a viewing behavior storage unit that builds a model of the viewing behavior by use of a Bayesian network having the viewing time period and the attribute as random variables and stores a conditional probability table of the Bayesian network; and a viewing behavior updating unit that updates the conditional probability table using the discretized viewing time periods.

Viewing behavior learning apparatus

A viewing recommendation apparatus includes an acquisition unit configured to acquire first broadcast program information for each of first broadcast programs to be broadcasted, a storage unit configured to store a plurality of previously broadcasted second broadcast programs in correspondence with second broadcast program information, a calculation unit configured to calculate an urgency in accordance with the first broadcast program information and the second broadcast program information to obtain a plurality of urgencies, the urgency indicating a degree to view a broadcast program earlier, and a generation unit configured to generate a recommendation list of programs to be viewed based on levels of the urgencies.

Viewing recommendation apparatus and method

A behavior prediction apparatus includes an input unit inputting sole behaviors of a human and simultaneous occurrence probability of behaviors, a behavior discrimination unit specifying a behavior pattern based on correspondence between behaviors corresponding to simultaneous occurrence probability and actual behaviors, an information generation/recording unit generating behavior history information of the specified behavior pattern, and to generate and record information of each of a behavior transition probability, a behavior time and a behavior occurrence probability based on the behavior history information, a behavior prediction unit obtaining a starting behavior from the behavior occurrence probability, select another behavior pattern in order of behavior occurrence probability, and add the behavior time to the selected behavior pattern to output behavior prediction information, and a social value calculation unit obtaining a social value per unit time using the behavior prediction information, social value unit information and the prediction period.

Behavior prediction apparatus and method therefor

According to an aspect of the present invention, there is provided with a behavior determining apparatus including: an elemental-behavior extracting unit configured to extract, from trajectory data in which positions of a traveling body in an observation area are held at predetermined time intervals, L (L is an integer equal to or larger than 1) elemental behavior quantities among a stopping time, a stop frequency, a walking time, a running time, a direction change frequency, a sum of angle change quantities at the times when the traveling body changes a traveling direction, a sum of absolute values of angle change quantities at the times when the traveling body changes a traveling direction, a gravitational center set vector, and a pre-stop average speed; and a behavior determining unit configured to determine a behavior of the traveling body from the L elemental behavior quantities.

Behavior recognition using vectors of motion properties based trajectory and movement type

A time series pattern generation unit extracts a key concept corresponding to a text of each data element of a plurality of time series data by referring to an extraction rule memory, and generates a plurality of time series patterns. A colony generation unit unifies a plurality of data elements having the same key concept and substantially the same time duration, and generates colonies of unified elements. In response to a selection of a first colony, a colony extraction unit specifies a second colony having a key concept contrary to the first colony, and extracts colonies earlier than the first colony and the second colony. A time reason extraction unit extracts a time reason as a divergence condition of the first colony from the second colony from the extracted colonies.

Time series pattern extraction apparatus and method

A method for displaying multidimensional data on a screen of a display device, includes assigning a first dimension element, having been selected by a user from among the multidimensional data, to an axis; giving weights to a plurality of member regions using a function of number of data belonging to respective members of the first dimension element; defining the weighted member regions as a plurality of regions on the axis, which are partitioned for the respective members of the first dimension element; arranging the weighted member regions within the screen to form a table; and displaying, in each cell region corresponding to the weighted member region, data belonging to each cell region and belonging to a second dimension element having been selected by the user from among the multidimensional data. The second dimension element is different from the first dimension element.

Multidimensional data display program

A time-series data analyzing apparatus which extracts a composite factor time-series pattern from time-series data. The apparatus includes a dividing device which divides the time-series data into pattern generation time-series data and pattern inspection time-series data which do not include pattern generation time-series data. A first generating device generates a transitional pattern including a support time data indicating a transition of support time and having a transition occurrence probability higher than a minimum occurrence probability in the pattern generation time-series data. A second generating device generates frequently appearing integrated transitional patterns. A second computing device computes cause-and-effect strength of each of the frequently appearing integrated transitional patterns using the pattern inspection time-series data. A display device displays the composite factor time-series pattern having the cause-and-effect strength higher than the minimum cause-and-effect strength given preliminarily.

Time-series data analyzing apparatus

Multidimensional data display apparatus, method, and multidimensional data display program

An analogical reasoning device comprises a knowledge base to store a group of knowledge data described in non-logic symbols and divided into more than one domains according to a certain division standard, an analogy detector to compare the knowledge data in one domain with the those in other domains stored in the knowledge base and determine similar domains containing the knowledge data which can be made identical by replacing a part of the non-logic symbols and an analogical reasoning unit to make analogy for a new knowledge data for the domains determined similar by the analogy detector.

Growing community of inventors

Tokyo, Japan

Ryohei Orihara