International Business Machines Corporation

Masaki Saitoh

Yoshio Horiuchi

Asuka Unno

Isao Kadowaki

Yuriko Nishikawa

Tomoko Murayama

Tsend Ochir Bat Ulzii

Masayuki Yamana

Yutaka Oishi

Takuya Goto

Munetaka Ohtani

Chiaki Oishi

Yusuke Nishitani

Masabumi Koinuma

Mari (Kuroki) Onishi

Shuji Umehara

Yoshinori Tahara

Chikafumi Yasumoto

Kohji Nakamori

Aspects for enabling private communications during a web conference are described. In an embodiment, one or more processors are configured for establishing a main channel for communication in a web conference comprising a plurality of users attending the web conference; transmitting main channel audio data to computing devices associated with the plurality of users; receiving a first sub-channel request comprising an invitation from a first user to at least a second user of the plurality of users; establishing a first sub-channel for communication in the web conference comprising the first user and at least the second user; generating first sub-channel audio data comprising reduced main channel audio data and first sub-channel input audio data comprising first user input audio data and second user input audio data; and transmitting the first sub-channel audio data to the first sub-channel.

Enabling private communications during a web conference

A computer-implemented method includes: selecting a web page from plural web pages being opened in a web browser; finding a similar page having similar content to the selected web page from the plural web pages; and closing the similar page in response to an operation by a user.

Restructuring pages having similar content

A method, computer system, and a computer program product for digital image recognition determination using a learned model is provided. The present invention may include acquiring a first determination result by making a determination concerning first data, using a first learned model. The present invention may include selecting a partial region of the first data. The present invention may then include generating second data obtained by applying a first alteration process to the partial region. The present invention may also include acquiring a second determination result by making a determination concerning the second data, using a second learned model. The present invention may lastly include obtaining a final determination result based on the first determination result and the second determination result.

Determination using learned model

A method, computer system, and a computer program product for predicting a variation of sequential blood glucose levels by using deep learning is provided. The present invention may include training a predictor associated with a user by using a deep learning network. The present invention may further include predicting a plurality of sequential blood glucose levels by the trained predictor based on at least one meal image, at least one time-period, and at least one set of data associated with a plurality of blood glucose levels of the user.

Estimating sequential blood-sugar levels using images of meals

Improved data pipelines are provided. A request to activate a first pipeline based on a first pipeline definition is received. The first pipeline is generated based on the first pipeline definition, wherein generating the first pipeline includes generating a reusable processing module at a first position within the first pipeline, and generating a downstream processing module at a second position within the first pipeline, where the downstream processing module selectively sends output data to the reusable processing module, such that data processed by the first data pipeline will be processed by the identified reusable processing module at least twice. The first pipeline is activated.

Establishing data pipelines with inclusion of iterative sub-pipelines

Hidden data co-occurrence relationships may be detected by a computer-implemented method, including monitoring data processing events on one or more server computers, gathering co-occurrences between a plurality of the data processing events, and generating one or more lineages between a plurality of directories associated with the plurality of the data processing events based on the gathered co-occurrences.

Detection of hidden data co-occurrence relationships

A first user input event initiating a transition of a GUI from a first state to a second state is generated in response to detecting a first user input received by the GUI. A first spatial location in the GUI selected by the first user input can be identified. A second user input event is generated in response to detecting a second user input received by the GUI while the GUI is in the second state. A second spatial location in the GUI selected by the second user input can be identified. A level of discontinuity between the first user input and the second user input can be determined. If the level of discontinuity does not exceed a threshold value, at least one user input event handling process can be selected to process the second user input event based on the level of discontinuity not exceeding the discontinuity threshold value.

Operation of a data processing system during graphical user interface transitions

A method, system and computer program product for assisting users to execute content in a user's computing environment. A user copying and pasting content from a first electronic document is detected, where the content is set as a first character string. Furthermore, the modification of the copied content is detected, where the modification includes modification of parameters of the user's computing environment to implement execution of the copied content and the modified copied content is set as a second character string. The first and second character strings as well as the modification points are registered in a repository. The modification points include the modified parameters of the first user's computing environment. By matching the character string copied by a second user from a second electronic document with one of the registered character strings, the second user may be presented with the modification points associated with the matching character string.

Assisting users to execute content copied from electronic document in user's computing environment

Provided is a method, wherein the system includes an existing pipeline having at least one processing module that is activated and information relating to the existing pipeline. The system operating method includes detecting a processing module to be shared that is shared with the existing pipeline, among the processing modules included in a new pipeline that is newly activated or in an activated pipeline that is already activated; and connecting a non-shared processing module that is not the processing module to be shared, among the processing modules included in the new pipeline or in the activated pipeline, to a shared processing module corresponding to the processing module to be shared in the existing pipeline.

Method for sharing processing modules between pipelines

Cloud services may be utilized by implementing an I/O hook that can intercept a data access from an application, migrating data accessed by the application to a cloud server that provides a cloud service, receiving, from the I/O hook, a data access intercepted from the application, and redirecting the intercepted data access to the cloud server.

Cloud service utilization

A computer-implemented method includes, by a first terminal device among a plurality of terminal devices in a vicinity of a user, receiving a frequency of coexistence from each of one or more second terminal devices among the plurality of terminal devices. The first terminal device and each of the second terminal devices each store a frequency of coexistence, wherein the frequency of coexistence shows a frequency that a terminal device which stores the frequency of coexistence and each of one or more other terminal devices coexisted in the vicinity at a predetermined location. The method further includes calculating a mean value of the frequencies of coexistence for each of the terminal devices which are not currently in the vicinity of the user, and, if the mean value is equal to or larger than a predetermined threshold, concluding that a terminal device is not in the vicinity of the user.

Wearable device detection

The present invention improves application of a style to a view object when a document for a Web page to be edited is edited on a browser-type edit screen. First, a view object is detected from a managed document. Then, a direct style directly described in the managed document and an indirect style identified only by referring to an external document are collected. A browser-type edit screen is generated in which the direct and indirect styles are applied to each view object. The content of the managed document is synchronized with the edited content on the browser-type edit screen based on the editing operations on the browser-type edit screen.

Web page authoring

Embodiments of the present invention provide a method, computer program product, and computer system for login on a wearable device. The method includes detecting one or more preregistered devices near a first wearable device. In response to determining that the detected one or more registered devices is greater than or equal to a predetermined number set forth in a simple password, the method requests the simple password from the user. If the correct simple password is entered, a first set of login information for the first wearable device is automatically inputted. The method may also include automatically inputting a second set of login information for one or more applications on the first wearable device.

Easy login on wearable device and wearable device applications

The present invention provides a Web page edit screen that allows an author to edit parent and child documents collectively while showing, in an appropriate manner, the content of the child document embedded in the parent document. Associated documents of a Web page to be edited are managed as managed documents individually. Document-by-document structure information is derived from the content of each managed document. Web page structure information is then assembled by connecting the document-by-document structure information based on embed-related code. Edit screen data is generated based on the content of the managed documents and the Web page structure information to generate a browser-type edit screen. The content of a relevant managed document is synchronized with the edited content on the edit screen. The edit screen data is matched with the edited content on the edit screen.

Growing community of inventors

Yokohama, Japan

Masaki Saitoh