Information retrieval using structured resources for paraphrase resolution
AssigneeInternational Business Machines Corporation
InventorsStephen A. Boxwell, Jared M. Smythe
Life of this patent
Estimating the term…
Abstract
An approach is provided in which a knowledge manager creates a pattern set based on training data that includes paraphrases and a set of first syntactic patterns. The knowledge manager receives a user question and matches one of the first syntactic patterns to a second syntactic pattern generated from the user question. Based on the matching, the knowledge manager generates new questions using the paraphrases in the pattern set and utilizes the new questions to query a second set of data and generate candidate answers that correspond to the user question.
Claims17
A method implemented by an information handling system that includes a memory ad a processor, the method comprising:
- retrieving a first set of training data and a second set of training data from a first set of resource data, wherein the first set of training data comprises a first question/answer pair and the second set of training data comprises a second question/answer pair;
- creating a first query, a first paraphrase, and a first syntactic pattern from the first question/answer pair, wherein the first syntactic pattern and the first paraphrase are based on one or more first syntactic properties of the first question/answer pair;
- creating a second query, a second paraphrase, and a second syntactic pattern from the second question/answer pair, wherein the second syntactic pattern and the second paraphrase are based on one or more second syntactic properties of the second question/answer pair;
- in response to determining that the first query matches the second query, adding the first syntactic pattern, the first paraphrase, the second syntactic pattern, and the second paraphrase to a pattern set;
- generating a question syntactic pattern based upon a user question;
- in response to correlating the question syntactic pattern to the first syntactic pattern in the pattern set, creating one or more new questions based on at least the second paraphrase stored in the pattern set;
- utilizing the one or more new questions to query a second set of resource data;
- and
- generating one or more candidate answers to the user question in response to querying the second set of resource data utilizing the one or more new questions.
An information handling system comprising:
- one or more processors;
- a memory coupled to at least one of the processors;
- and
- a set of computer program instructions stored in the memory and executed by at least one of the processors in order to perform actions of:
- retrieving a first set of training data and a second set of training data from a first set of resource data, wherein the first set of training data comprises a first question/answer pair and the second set of training data comprises a second question/answer pair;
- creating a first query, a first paraphrase, and a first syntactic pattern from the first question/answer pair, wherein the first syntactic pattern and the first paraphrase are based on one or more first syntactic properties of the first question/answer pair;
- creating a second query, a second paraphrase, and a second syntactic pattern from the second question/answer pair, wherein the second syntactic pattern and the second paraphrase are based on one or more second syntactic properties of the second question/answer pair;
- in response to determining that the first query matches the second query, adding the first syntactic pattern, the first paraphrase, the second syntactic pattern, and the second paraphrase to a pattern set;
- generating a question syntactic pattern based upon a user question;
- in response to correlating the question syntactic pattern to the first syntactic pattern in the pattern set, creating one or more new questions based on at least the second paraphrase stored in the pattern set;
- utilizing the one or more new questions to query a second set of resource data;
- and
- generating one or more candidate answers to the user question in response to querying the second set of resource data utilizing the one or more new questions.
A computer program product stored in a computer readable storage medium, comprising computer program code that, when executed by an information handling system, causes the information handling system to perform actions comprising:
- retrieving a first set of training data and a second set of training data from a first set of resource data, wherein the first set of training data comprises a first question/answer pair and the second set of training data comprises a second question/answer pair;
- creating a first query, a first paraphrase, and a first syntactic pattern from the first question/answer pair, wherein the first syntactic pattern and the first paraphrase are based on one or more first syntactic properties of the first question/answer pair;
- creating a second query, a second paraphrase, and a second syntactic pattern from the second question/answer pair, wherein the second syntactic pattern and the second paraphrase are based on one or more second syntactic properties of the second question/answer pair;
- in response to determining that the first query matches the second query, adding the first syntactic pattern, the first paraphrase, the second syntactic pattern, and the second paraphrase to a pattern set;
- generating a question syntactic pattern based upon a user question;
- in response to correlating the question syntactic pattern to the first syntactic pattern in the pattern set, creating one or more new questions based on at least the second paraphrase stored in the pattern set;
- utilizing the one or more new questions to query a second set of resource data;
- and
- generating one or more candidate answers to the user question in response to querying the second set of resource data utilizing the one or more new questions.
Dependencies are inferred from the claim text.