System and method for augmenting answers from a QA system with additional temporal and geographic information

A computer program product for generating answers to natural language questions, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:

• receive a first natural language question;
• decompose the first natural language question into at least one database query;
• generate, based at least in part on a text passage, a first answer in a first answer set corresponding to the first natural language question, wherein the text passage does not include the first natural language question, and wherein the text passage provides evidence supporting the first answer;
• mine the text passage to identify at least one modifier selected from the group consisting of a geospatial modifier in the text passage and a temporal modifier in the text passage;
• generate a second natural language question based on the first natural language question and the at least one modifier;
• generate a second answer set corresponding to the second natural language question, a second answer in the second answer set corresponding to the first answer, wherein each answer of the second answer set includes the at least one modifier;
• generate a confidence score for each answer in the first answer set and each answer in the second answer set, wherein the confidence score for the first answer in the first answer set is based on a first number of documents in which the first answer appears and the confidence score for the second answer in the second answer set is based on a second number of documents in which the second answer appears;
• and
• return a question answer set comprising the first answer set, the second answer set, or a combination of the first answer set and the second answer set to a user.

A computer-implemented question answering system, comprising:

• a memory having instructions therein;
• and
• at least one processor in communication with the memory, wherein the at least one processor is configured to execute the instructions to:
• receive a first natural language question;
• decompose the first natural language question into at least one database query;
• generate, based at least in part on a text passage, a first answer in a first answer set corresponding to the first natural language question, wherein the passage does not include the first natural language question, and wherein the text passage provides evidence supporting the first answer;
• mine the text passage to identify at least one modifier selected from the group consisting of a geospatial modifier in the text passage and a temporal modifier in the text passage;
• generate a second natural language question based on the first natural language question and the at least one modifier;
• generate a second answer set corresponding to the second natural language question, a second answer in the second answer set corresponding to the first answer, wherein each answer of the second answer set includes the at least one modifier;
• generate a confidence score for each answer in the first answer set and each answer in the second answer set, wherein the confidence score for the first answer in the first answer set is based on a first number of documents in which the first answer appears and the confidence score for the second answer in the second answer set is based on a second number of documents in which the second answer appears;
• and
• return a question answer set comprising the first answer set, the second answer set, or a combination of the first answer set and the second answer set to a user.