International Business Machines Corporation

Chao Yuan Huang

Yi-Chun Chen

Der-Joung Wang

Wen-Ping Chi

Yi-Lin Tsai

Yen-Min Wu

Hsiao-Yung Chen

Yin Qian

Lin Chung Liang

Ting-Yi Wang

Yen Lin Li

Chi-Ying Chang

Pei-Yi Lin

Fei Shen

Jin Shi

David S C Chen

Hung-Jung Lai

Jen Ping Cheng

Po-Chun Lin

Micky W T Chiang

Chia-Hsueh Lin

Fan Yang

Yu Gu

Xiao Ming Ma

Yan Wu

Chih-Yuan Lin

Yu-Hsing Wu

Albert Hsieh

Sharon Chen

Wen Ke Xue

Li-Chuan Yen

Shu-chih Chen

David Shao Chung Chen

Yuan Jie Zhang

Qiang Song

Xiang Yu Song

Hsinchi Chang

Jacqueline Cheng

An-Ting Tsai

Micky Wei-Te Chiang

Kee Hooi Yeow

Po Chun Lin

Peng Chen

Chia Hsin Chuang

Shun Jie Yu

Yu Jie Gu

A method, computer system, and a computer program product for task assistance is provided. The present invention may include acquiring a request expression input by a user. The present invention may include identifying a request intent associated with a task based on the request expression. The present invention may include determining a response script corresponding to the request intent. The present invention may include executing the response script to complete the task and presenting the process of running the task in a user-interface (UI).

Script-based task assistance

From an input image comprising a portion of text, the portion of text is extracted. The portion of text comprises text in a first natural language. The input image is classified into a candidate scene within a set of predefined scenes. The candidate scene is scored according to a quality measure. Using a predefined scene parameter of the scored candidate scene, the portion of text is translated. In the input image, the portion of text is replaced. with the translated portion of text.

Scene recognition based natural language translation

One or more computer processors identify a string within source code associated with a program integrated interface (PII). The one or more computer processors compare the identified string to a string entry comprised in a PII dictionary. The one or more computer processors, responsive to a match between the identified string and the string entry in the PII dictionary, classify the identified string as translatable or non-translatable based on a classification associated with the string entry. The one or more computer processors, responsive to a non-match between the identified string and the string entry in the PII dictionary, classify the identified string as translatable or non-translatable utilizing a hyperplane model trained with a margin (C) derived from unlimited search. The one or more computer processors, responsive to a classified translatable string, translate the classified translatable string.

Hardcoded string detection

Embodiments of the present disclosure relate to attribute-based selectable item recommendation. A computer-implemented method is disclosed. A computer generates respective attribute profiles of selectable items based on one or more attribute elements included in the selectable items. The profiles include at least one attribute elements respectively correspond to the selectable items. The computer generates an attribute requirement description of an item requesting agent based on a historical search dataset of the item requesting agent. The attribute requirement description includes at least part of the one or more attribute elements. The computer determines recommended selectable items included in the selectable items for the item requesting agent based on the attribute profiles of the selectable items and the attribute requirement description of the item requesting agent. The attribute profiles of the recommended selectable items are suitable with the attribute requirement description based on the one or more attribute elements.

Item recommendation based on computed similarity between automatically generated item and request profiles

Embodiments are disclosed for a method for dynamic text reading. The method includes performing pre-processing for a text document. Pre-processing includes determining the text document comprises an emotional statement based on an indicator of an emotion associated with the emotional statement. Pre-processing also includes identifying a speaker of the emotional statement. Further, pre-processing includes generating a role-to-voice map that associates the speaker with a digital representation of a voice for the speaker. The method additionally includes generating, based on the pre-processing, the voice for the speaker reading aloud a text of the text document using the digital representation of the voice with a tonal modulation that conveys the emotion.

Dynamic text reader for a text document, emotion, and speaker

A method, computer system, and a computer program product for securing message transmission is provided. The present invention may include linking, by a first terminal device, a communication interface to a first communication channel for a target application. The present invention may include, in response to a first message being inputted in the communication interface, encrypting, by the first terminal device, the first message with a key for the first communication channel. The present invention may include, transmitting, by the first terminal device, the encrypted first message via the first communication channel.

Message security

Orchestrated chat services utilizing a centralized chat service with access to a plurality of chatbots registered to the chat service. User's seeking support from the orchestrated chat service interact with a single user interface, while the backend of the chat service extracts the intents and entities from the user's input into the chat service. The orchestrated chat service identifies one or more classifications of chatbots suitable for responding to the user's input within a prescribed level of confidence dictated by one or more orchestration rules and selects a chatbot predicted to most likely respond to the user's input in a correct and accurate manner. The orchestrated chat service formats the user input and chat history into format of the selected chatbot's API, forwards user input and history to the selected chatbot and returns the response from the selected chatbot to the user interface of the orchestrated chat service.

Chatbot orchestration

The present disclosure provides a method, computer system and computer program product for determining interruption points based on emotion values in a content. According to the method, emotion values of one or more emotion catalogs at one or more time points of the content to be interrupted can be determined, and one or more interruption points can be determined based on the determined emotion values.

Interruption point determination

Aspects of the present disclosure relate to recommendation and translation of symbols. In some embodiments, the method includes receiving current text data from a sender, receiving a first set of historical text data for the sender, determining emoji use preference of the sender, analyzing the current text data for context that is representable by an emoji, selecting a first emoji based the context and the first set of historical text data, receiving a second set of historical text data for a recipient of the current text data, analyzing a contextual meaning of the selected emoji based on the first set of historical text data and the second set of historical text data, determining, based on the analyzing the contextual meaning of the selected emoji, that a second emoji would more closely match the context of the current text data, and sending the second emoji to the recipient.

Recommendation and translation of symbols

A method and system of authenticating a user are provided. A request for a resource is received by a server, from a user device. A predefined number is received from the user device. A first number and a second number are created. The first number is sent to the user device. A first discrete logarithm is determined based on a challenge code and the first number and sent to the user device. A first pass code is calculated via a second discrete logarithm based on the first discrete logarithm, the predefined number, and the first number. A second pass code based on the second discrete logarithm, is received from the user device. The first pass code is compared to the second pass code. Upon determining that the first pass code is identical to the second pass code, the user device is allowed access a resource associated with the computing device.

Multi factor authentication

A method and system of authenticating a user are provided. A request for a resource is received from a user device. A predefined number is received. A first number and a second number are created. A first discrete logarithm based on the first number and the predefined number is determined and sent to the user device, together with the second number. A second discrete logarithm is received from the user device. A first pass code is calculated via a third discrete logarithm, based on the second discrete logarithm, the first number, and the predefined number. A second pass code is received via a fourth discrete logarithm, based on the first discrete logarithm, the third number, and the predefined number. Upon determining that the first pass code is identical to the second pass code, the user device is allowed to access a resource associated with the computing device.

Disclosed are a method and a computer system for splitting and merging files via a motion input on a graphical user interface. The method comprises determining existence of a motion input; splitting the file into split multiple partial files based on pre-configurations; and generating multiple partial file icons representing the split multiple partial files on the graphical user interface. The multiple partial files may be subsequently merged together.

Splitting and merging files via a motion input on a graphical user interface

A method for context-aware translation is provided. The present invention may include receiving a program integrated information file (PII) associated with a graphical user interface object of with a software product. The present invention may also include indexing each element string within the PII file. The present invention may further include generating a base language general availability build of the graphical user interface object using the PII file. The present invention may also include retrieving a document object model (DOM) of each element within the base language general availability build. The present invention may further include creating a string relation table for the base language general availability build using the DOM for each element and each indexed element string within the PII file. The present invention may also include generating a string relation translation memory table for at least one language.

Context-aware translation memory to facilitate more accurate translation

A computer-implemented method includes identifying a text segment including one or more text artefacts and determining a language-specific occurrence database associated with the text segment. The computer-implemented method further includes, for each of said one or more text artefacts, determining a language-specific occurrence value based on the language-specific occurrence database and determining a language-specific compliance value based on each language-specific occurrence value. A corresponding computer program product and computer system are also disclosed.

Identifying corrupted text segments

Growing community of inventors

Taipei, Taiwan

Chao Yuan Huang