Abstract: A natural language programming system may configure a machine learning (ML) model to translate natural language descriptions into constrained language statements. The constrained language statements may express the natural language descriptions using a constrained subset of natural language. The constrained subset of natural language includes words with unambiguous semantics and with meaning that has a clear and singular interpretation. The constrained language statements with unambiguous semantics enable construction of valid statements in high-level “English-like” executable programming language. With the present system, a user does not need to learn a new programming language but rather learn to constrain their natural language statements to a subset of the natural language (“constrained language”) and to generate executable programs.

Abstract: A natural language programming system may configure a machine learning (ML) model to translate natural language descriptions into constrained language statements. The constrained language statements may express the natural language descriptions using a constrained subset of natural language. The constrained subset of natural language includes words with unambiguous semantics and with meaning that has a clear and singular interpretation. The constrained language statements with unambiguous semantics enable construction of valid statements in high-level “English-like” executable programming language. With the present system, a user does not need to learn a new programming language but rather learn to constrain their natural language statements to a subset of the natural language (“constrained language”) and to generate executable programs.

Abstract: A collaborative user support system may include support components and a user portal to interact with a user and generate solutions to user problems. The support components may include a natural language understanding (NLU) engine, a reasoning engine, a semantic search engine, a multimodal dialog engine, and an active learning engine. The user portal may receive input dialog and the NLU engine may translate the dialog into machine language. The reasoning engine may determine whether the dialog includes a problem and call the semantic search engine to identify a potential solution. The multimodal dialog engine may determine a visual representation of a problem and/or solution. The active learning engine may receive user feedback on generated solutions for continuous improvements to the system.

Abstract: A device for charging number of cells connected together in a row and create a battery voltage, the charging is done in such a way that every single cell in a battery (or several small cells together) is charged separately by the computer control and moves automatically to the next cell until all cells comprising the battery are full.