DETECTING AND EXECUTING DATA RE-INGESTION TO IMPROVE ACCURACY IN A NLP SYSTEM

• US 20140280253A1
• Filed: 03/14/2013
• Published: 09/18/2014
• Est. Priority Date: 03/12/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Clark, Adam T., Dubbels, Joel C., Huebert, Jeffrey K., Petri, John E.
• Granted Patent: US 9,245,009 B2
• Time in Patent Office: Days
• US Class Current: 707/755
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/24552 Database cache management
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/9574 of access to content, e.g. by caching

Abstract

In some NLP systems, queries are compared to different data sources stored in a corpus to provide an answer to the query. However, the best data sources for answering the query may not currently be contained within the corpus or the data sources in the corpus may contain stale data that provides an inaccurate answer. When receiving a query, the NLP system may evaluate the query to identify a data source that is likely to contain an answer to the query. If the data source is not currently contained within the corpus, the NLP system may ingest the data source. If the data source is already within the corpus, however, the NLP may determine a time-sensitivity value associated with at least some portion of the query. This value may then be used to determine whether the data source should be re-ingested—e.g., the information contained in the corpus is stale.

Claims

1. A method comprising:

receiving a query for processing by a natural language processing (NLP) system;

identifying a data source related to the query by associating one or more elements of the query to the data source;

upon determining that the related data source is not in a corpus of the NLP system, ingesting the related data source into the corpus; and

upon determining that the related data source is in the corpus of the NLP system;

determining a time-sensitivity value associated with the query indicating a degree to which an accurate answer to the query is dependent on a staleness of the related data source, and

upon determining that the time-sensitivity value satisfies a staleness threshold, re-ingesting the related data source into the corpus.

2. The method of claim 1, further comprising:

performing a NLP technique to compare the query to the corpus to identify at least one answer to the query after one of (i) ingesting the related data source into the corpus and (ii) re-ingesting the related data source into the corpus; and

transmitting the at least one answer to an entity submitting the query.

3. The method of claim 1, wherein identifying the related data source further comprises:

parsing the query to identify an element of the query that specifically provides the data source;

characterizing at least one element in the query by assigning the at least one element a topic; and

identifying the related by comparing the topic to a plurality of topics associated with different data sources.

4. The method of claim 1, wherein determining the time-sensitivity value associated with the query further comprises:

performing a concept mapping to assign a concept to at least one element in the query;

upon determining that the concept is related to duration or time, assigning the time-sensitivity value based on duration or time specified in the concept; and

upon determining that the concept is dependent on time by matching the concept to a pre-defined list of time-dependent concept, assigning the time-sensitivity value based on the time dependency of the concept.

5. The method of claim 1, wherein determining that the time-sensitivity value satisfies the staleness threshold, further comprises determining whether the related data source is stale by comparing a time-stamp associated with the related data source to the time-sensitivity value.

6. The method of claim 1, further comprising, before ingesting or re-ingesting the data source, receiving permission from a user submitting the query to ingest or re-ingest the data source into the corpus.

7. The method of claim 1, wherein the corpus comprises a plurality of data from different data sources, wherein the data of the different data sources are organized based on a common format in the corpus.