SBIR-STTR Award

LVI proposes developing an advanced framework of lifted probabilistic inference algorithms for enhancing the scaling and accuracy of text analytics. In Phase I of this project, LVI proposes to explore the scalability of various lifted inference techniques for utilizing Markov Logic Networks (MLN) and the Tractable Markov Logic language in the Tuffy software package as a base platform. These techniques will be applied to the area of automated knowledge base construction from free text, using abductive reasoning to infer the optimal updates to the knowledge base. The MLN will use unsupervised joint inferencing techniques to combine record segmentation, co-reference resolution, and entity resolution in a single process, as opposed to a pipelined approach. Phase I will also include a cost-benefit analysis of utilizing ontologies as background knowledge as a method of bootstrapping MLN learning, using general-purpose models from Linked Open Data (LOD) Cloud sources such as DBpedia and YAGO. LVI's proposed solution is based on the company's extensive experience in developing tools utilizing Semantic Web based AI technologies, inference engines, and Linked Data applications. LVI brings deep experience in knowledge engineering, information extraction, data integration through statistical ontology alignment, machine learning, and online analytical processing (OLAP), and intelligent application development.

Benefit:
The algorithms for Lifted inference have a variety of applications. They include Social networks, object recognition, link prediction, activity recognition, model counting, bio-medical applications and relational reasoning and learning. Fundamental building block to improve current capabilities.

Keywords:
Joint & Lifted Inference, Belief Propagation, First Order Logic, Probabilistic Inference, Text Analytics, Natural Language Processing, Nlp, Statistical Relational Learning

LVI proposes developing an advanced framework of lifted probabilistic inference algorithms for enhancing the scaling and accuracy of text analytics. In Phase I, LVI explored the scalability of various lifted inference techniques for utilizing Markov Logic Networks (MLN) in the Tuffy software package. Phase I also included investigation and demonstration of DeepDive, a scalable, high-performance inference and learning engine for text analytics. These techniques were applied for automated knowledge base construction from free text, using abductive reasoning for optimal updates to the knowledge base. The MLN used unsupervised joint inferencing to combine record segmentation, co-reference resolution, and entity resolution in a single process, as opposed to a pipelined approach. The Phase II end-to-end prototype will be developed to perform text analytics over an information repository using the optimized joint inference technique. The prototype capabilities including joint inference over cross-document and multiple knowledge bases will be demonstrated through, for example, answering specific queries without considering the entire model and/or the entire evidence. "Distance supervision" and the Stanford Dependency Parser for NLP will be used to leverage external data sources for entity identification. Collaborating with a large financial institution MLN will be developed for entity recognition, relationship discovery and classification.

Benefits:
The algorithms for Lifted inference have a variety of applications. They include Social networks, object recognition, link prediction, activity recognition, model counting, bio-medical applications and relational reasoning and learning. Fundamental building block to improve current capabilities.

Keywords:
Joint & lifted inference, belief propagation, first order logic, probabilistic inference, text analytics, natural language processing, NLP, Statistical Relational Learning