Determining correct answers to tax and accounting issues arising from business transactions and generating accounting entries to record those transactions using a computerized predicate logic implementation

A non-transitory computer readable medium comprising computer executable instructions which when executed by a computer cause the computer to perform the method of:

• paraphrasing a plurality of pertinent authority documents, stored in ordinary English language in an electronic database, into a plurality of series of intermediate forms, by text processing with a trained paraphrasing machine learning module, executing on at least one hardware processor, in accordance with rules obtained at least in part from a paraphrasing training process, said paraphrasing training process encoding into said rules natural language interpretation semantics from judicial decisions interpreting said authority documents, said plurality of pertinent authority documents each comprising at least one of a tax authority and an accounting authority;
• rendering said plurality of series of intermediate forms into a corresponding plurality of series of logical statements in predicate logic extended (PLE), using a trained intermediate form-logic machine learning module, executing on said at least one hardware processor, in accordance with rules obtained at least in part from an intermediate form-logic training process;
• encoding said plurality of series of logical statements in predicate logic extended (PLE) into a plurality of relations in a logic programming language by processing said logical statements with a PYTHON program executing on said at least one hardware processor to obtain said relations in said logic programming language, said logic programming language embodying a programming paradigm based on formal logic, said relations in said logic programming language comprising a set of sentences in logical form, expressing facts and rules about a corresponding pertinent problem domain;
• indexing said plurality of pertinent authority documents in said electronic database to facilitate subsequent searching to locate at least one of said plurality of pertinent authority documents applicable in a given situation;
• obtaining, via a user interface, at least one of user-entered query terms and query terms derived from pertinent prose, at least one of said user-entered query terms and query terms derived from pertinent prose specifying at least one of a tax problem and an accounting problem;
• based on said at least one of user-entered query terms and query terms derived from pertinent prose, searching said electronic database to locate said at least one of said plurality of pertinent authority documents applicable in said given situation;
• responsive to said searching and locating, activating a comparison and inference engine executing on said at least one hardware processor, to apply those of said plurality of relations in said logic programming language corresponding to said at least one of said plurality of pertinent authority documents applicable in said given situation to a problem ascertainable from said at least one of user-entered query terms and query terms derived from pertinent prose, to obtain a conclusion by applying predicate derivation extended (PDE);
• and
• presenting said conclusion to a decision maker, via said user interface.

An apparatus comprising:

• a memory;
• at least one processor, coupled to said memory;
• and
• a non-transitory computer readable medium comprising computer executable instructions, said computer executable instructions including a trained paraphrasing machine learning module, a trained intermediate form-logic machine learning module, a PYTHON program, and a comparison and inference engine, said computer executable instructions, when loaded into said memory, configuring said at least one processor to be operative to:
• paraphrase a plurality of pertinent authority documents, stored in ordinary English language in an electronic database, into a plurality of series of intermediate forms, by text processing with said trained paraphrasing machine learning module, executing on said at least one processor, in accordance with rules obtained at least in part from a paraphrasing training process, said paraphrasing training process encoding into said rules natural language interpretation semantics from judicial decisions interpreting said authority documents, said plurality of pertinent authority documents each comprising at least one of a tax authority and an accounting authority;
• render said plurality of series of intermediate forms into a corresponding plurality of series of logical statements in predicate logic extended (PLE), using a trained intermediate form-logic machine learning module, executing on said at least one hardware processor, in accordance with rules obtained at least in part from an intermediate form-logic training process;
• encode said plurality of series of logical statements in predicate logic extended (PLE) into a plurality of relations in a logic programming language by processing said logical statements with said PYTHON program executing on said at least one processor to obtain said relations in said logic programming language, said logic programming language embodying a programming paradigm based on formal logic, said relations in said logic programming language comprising a set of sentences in logical form, expressing facts and rules about a corresponding pertinent problem domain;
• index said plurality of pertinent authority documents in said electronic database to facilitate subsequent searching to locate at least one of said plurality of pertinent authority documents applicable in a given situation;
• obtain, via a user interface, at least one of user-entered query terms and query terms derived from pertinent prose, at least one of said user-entered query terms and query terms derived from pertinent prose specifying at least one of a tax problem and an accounting problem;
• based on said at least one of user-entered query terms and query terms derived from pertinent prose, search said electronic database to locate said at least one of said plurality of pertinent authority documents applicable in said given situation;
• responsive to said searching and locating, activating said comparison and inference engine executing on said at least one hardware processor, to apply those of said plurality of relations in said logic programming language corresponding to said at least one of said plurality of pertinent authority documents applicable in said given situation to a problem ascertainable from said at least one of user-entered query terms and query terms derived from pertinent prose, to obtain a conclusion by applying predicate derivation extended (PDE);
• and
• present said conclusion to a decision maker, via said user interface.