System and method for question answering with derived glossary clusters

A system comprising:

• a processor;
• a data bus coupled to the processor;
• and
• a non-transitory computer-usable medium embodying computer program code, the computer-usable medium being coupled to the data bus, the computer program code used for answering general background questions on a topic from documents with glossary sections executable by the processor and configured for:
• receiving a set of documents with glossaries;
• extracting a set of terms T(T1, T2, . . . , Tn) and associated glossary entries G(G1, G2, . . . , Gn) wherein each term Ti has a corresponding glossary entry Gi;
• associating related glossary entries together based on a similarity algorithm to form glossary clusters C(C1, C2, . . . , Cn) wherein each glossary cluster Ci refers to one or more glossary entry Gi;
• receiving a query tailored to general information with query terms;
• ranking the glossary clusters C(C1, C2, . . . , Cn) based on a relevance to the query terms to form a ranked set;
• selecting a set of glossary clusters meeting a high ranked threshold;
• and
• providing the general information based on extracting information from the glossary clusters that meet the high ranked threshold.

A non-transitory, computer-readable storage medium embodying computer program code, the computer program code comprising computer executable instructions configured for:

• receiving a set of documents with glossaries;
• extracting a set of terms T(T1, T2, . . . , Tn) and associated glossary entries G(G1, G2, . . . , Gn) wherein each term Ti has a corresponding glossary entry Gi;
• associating related glossary entries together based on a similarity algorithm to form glossary clusters C(C1, C2, . . . , Cn) wherein each glossary cluster Ci refers to one or more glossary entry Gi;
• receiving a query tailored to general information with query terms;
• ranking the glossary clusters C(C1, C2, . . . , Cn) based on a relevance to the query terms to form a ranked set
• selecting a set of glossary clusters meeting a high ranked threshold;
• and
• providing the general information based on extracting information from the glossary clusters that meet the high ranked threshold.