Identifying logic problems in text using a statistical approach and natural language processing

A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on at least one processor of a data processing system, causes the data processing system to implement a logic problem system, comprising a logic problem detection engine and a logic problem solution engine, for identifying a logic problem to be solved, wherein the computer readable program causes the at least one processor to:

• parse, by a parser executing on the at least one processor of the data processing system, the input text to identify entities in an input text received by the logic problem system by applying parser rules in a rules data structure;
• determine, by a statistical analysis component executing on the at least one processor of the data processing system, statistical relationships between pairs of entities in the input text to form a set of statistical relationships;
• determine, by a logic problem detection engine executing on the at least one processor of the data processing system, whether the set of statistical relationships includes a high density of standard relationships and one or more outlier relationships;
• responsive to determining that the set of statistical relationships includes a high density of standard relationships and one or more outlier relationships, present, by the logic problem detection engine, an indication that the input text includes a logic problem to a user;
• identify, by the logic problem detection engine, a question, missing fact, or statistical anomaly in the input text representing a goal of solving the logic problem;
• and
• send, by the logic problem detection engine to the logic problem solution engine, the input text, wherein the logic problem solution engine generates a plurality of candidate solutions to the logic problem ranked by confidence value.

An apparatus comprising:

• at least one processor;
• and
• a memory coupled to the at least one processor, wherein the memory comprises instructions which, when executed by the at least one processor, cause the at least one processor to implement a logic problem system, comprising a logic problem detection engine and a logic problem solution engine, for identifying a logic problem to be solved, wherein the instructions cause the at least one processor to:
• parse, by a parser executing on the at least one processor of the data processing system, the input text to identify entities in the input text;
• determine, by a statistical analysis component executing on the at least one processor of the data processing system, statistical relationships between pairs of entities in the input text to form a set of statistical relationships;
• determine, by the logic problem detection engine executing on the at least one processor of the data processing system, whether the set of statistical relationships includes a high density of standard relationships and one or more outlier relationships;
• responsive to determining that the set of statistical relationships includes a high density of standard relationships and one or more outlier relationships, present, by the logic problem detection engine, an indication that the input text includes a logic problem to a user;
• identify, by the logic problem detection engine, a question, missing fact, or statistical anomaly in the input text representing a goal of solving the logic problem;
• and
• send, by the logic problem detection engine to the logic problem solution engine, the input text, wherein the logic problem solution engine generates a plurality of candidate solutions to the logic problem ranked by confidence value.