International Business Machines Corporation

Zi Ming Huang

Jian Wang

Jing Li

Jian Min Jiang

Qi Cheng Li

Ke Fei Wang

Ya Bin Dang

Lijun Mei

Xue Han

Xin Ni

Jie Ma

Xin Zhou

Zhen Zhang

Christopher Jonathan Davis

Rachel Mohammed

Zhuoxuan Jiang

Wanqing Liu

A processor obtains a pair including a question and an answer in natural language; determines at least one entity in the question and an entity type of each of the at least one entity consistent with the schema of a knowledge graph (KG); identifies a subset of candidate query templates based on the entity type of each of the at least one entity, wherein the candidate query templates are generated based on the schema of the KG; composes a set of queries by populating the at least one entity into each of the subset of candidate query templates; executes the set of queries on the KG to generate respective answers; identifies a first answer from the respective answers that is matching with the answer in the pair; and determines a candidate query template, from the subset of candidate query templates, corresponding to the first answer as a query template.

Generation of query templates for knowledge-graph based question answering system

New question and answer (QA) pairs can be automatically discovered from a corpus of data such as online chats and conversations. Newly discovered QA pairs can augment QA database, which can be used by a computer processor or device, e.g., by a chatbot, an automated machine, and/or another. Existing QA knowledge can be used to learn the structures of QA knowledge distribution in conversations, and new QA knowledge can be automatically learned through the structure of learned QA knowledge distribution in conversations. The structure of learned QA knowledge distribution can be refined by adding more semantics based on labeled data.

Discovering new question and answer knowledge from conversation

Techniques for generating a natural language question template for an artificial intelligence question and answer (QA) system are disclosed. A graph database query relating to a QA system is parsed using a predefined schema. The parsing includes extracting a first plurality of values from the graph database query relating to a where clause in the graph database query, extracting a second plurality of values from the graph database query relating to a return clause in the graph database query, identifying a QA template rule relating to the graph database query, based on a match clause in the graph database query. A natural language question template is generated based on the first plurality of values, the second plurality of values, and the identified QA template rule. The natural language question template is suitable for use by the QA system as part of generating a response to a natural language question.

Generating question templates in a knowledge-graph based question and answer system

Methods, systems and computer program products for providing automated robotic process automation design generation are provided. Aspects include receiving a set of stored automated sub-processes. Aspects also include receiving a process description document that describes a desired computer automated task (CAT) process. Aspects also include identifying a plurality of candidate sub-processes based on the process description document. Aspects also include identifying at least one matching sub-process based on a comparison of each of the plurality of candidate sub-processes to the set of stored automated sub-processes. The at least one matching sub-process is a stored automated sub-process that exceeds a threshold level of similarity to a candidate sub-process. Aspects also include generating a recommendation to automatically populate a portion of a CAT design file for the desired CAT process based on the at least one matching sub-process.

Automated generation of robotic computer program code

In embodiments, a reusable and adaptive multi-task orchestration dialogue system orchestrates a set of single-task dialogue systems to provide multi-scenario dialogue processing. In embodiments, for each question propounded by a user, using a deep learning predictive model, a best single-task dialogue system is chosen out of the set. In embodiments, multi-task orchestration is done without the need to change, or even understand, the inner workings or mechanisms of the individual single-task dialogue systems in the set. Moreover, the multi-task orchestration is also unconcerned with what rules are set in each individual single-task dialogue system. In embodiments, prior to selection of the best single-task dialogue system to return the best answer, new intents and entities are discovered and used to update an existing dialogue path. In embodiments, additional data is continually collected, and used to retrain model so as to further improve performance.

Cognitive orchestration of multi-task dialogue system

A method and apparatus for providing power includes a power socket module having a first conducting layer and a second conducting layer. An insulating layer may be positioned in between the first and second conducting layers. A plug includes a first prong having a first length and second prong having a second length, where the first length is longer than the second length. When the plug is plugged into the power socket module, the first prong electrically couples to the second conducting layer, and the second prong electrically couples to the first conducting layer.

Power socket module and plug

Growing community of inventors

Beijing, China

Zi Ming Huang