Domain similarity scores for information retrieval
AssigneeSap SE
InventorsHans-Martin Ramsl, Sebastian Koehler
Life of this patent
Estimating the term…
Abstract
Various embodiments of systems, computer program products, and methods to provide domain similarity scores for information retrieval are described herein. In an aspect, a plurality of files associated with a plurality of domains are retrieved. A corpus corresponding to the plurality of domains is generated based on the plurality of files by integrating the plurality of files corresponding to the plurality of domains. Further, similarity scores between the plurality of domains are determined based on the generated corpus. The similarity scores between the plurality of domains are persisted in a similarity scores repository to enable information retrieval during translating data between different languages.
Claims20
A non-transitory computer-readable storage medium storing instructions, which when executed by a computer cause the computer to:
- retrieve a plurality of files associated with a plurality of domains;
- generate a corpus including a plurality of integrated domain files corresponding to the plurality of domains based on integrating the plurality of files associated with the plurality of domains;
- determine similarity scores between the plurality of domains based on the generated corpus, wherein a similarity score from the similarity scores between a first domain and a second domain from the plurality of domains is determined based on a geometric distance between the two domains in a vector space defined by corresponding two integrated domain files from the corpus;
- persist the similarity scores between the plurality of domains in a similarity scores repository to enable information retrieval during translating data between different languages;
- and
- when there is no data available for translating received data in a requested domain, search into a different domain from the plurality of domains associated with high similarity score to the requested domain, according to the persisted similarity scores, to provide a semantically coherent translation for the received data with the requested domain.
A computer implemented method to provide domain similarity scores for information retrieval, comprising:
- retrieving a plurality of files associated with a plurality of domains;
- generating a corpus including a plurality of integrated domain files corresponding to the plurality of domains based on integrating the plurality of files associated with the plurality of domains;
- determining the similarity scores between the plurality of domains based on the generated corpus, wherein a similarity score from the similarity scores between a first domain and a second domain from the plurality of domains is determined based on a geometric distance between the two domains in a vector space defined by corresponding two integrated domain files from the corpus;
- persisting the similarity scores between the plurality of domains in a repository to enable information retrieval during translating data between different languages;
- and
- when there is no data available for translating received data in a requested domain, searching into a different domain from the plurality of domains associated with high similarity score to the requested domain, according to the persisted similarity scores, to provide a semantically coherent translation for the received data with the requested domain.
A computing system comprising:
- at least one processor;
- and
- one or more memory devices communicative with the at least one processor, wherein the one or more memory devices store instructions to cause the at least one processor to:
- retrieve a plurality of files associated with a plurality of domains:
- generate a corpus including a plurality of integrated domain files corresponding to the plurality of domains based on integrating the plurality of files associated with the plurality of domains;
- determine similarity scores between the plurality of domains based on the generated corpus, wherein a similarity score from the similarity scores between a first domain and a second domain from the plurality of domains is determined based on a geometric distance between the two domains in a vector space defined by corresponding two integrated domain files from the corpus;
- persist the similarity scores between the plurality of domains in a repository to enable information retrieval during translating data between different languages;
- and
- when there is no data available for translating received data in a requested domain, search into a different domain from the plurality of domains associated with high similarity score to the requested domain, according to the persisted similarity scores, to provide a semantically coherent translation for the received data with the requested domain.
Dependencies are inferred from the claim text.