Ni-Square Analytics Pvt Ltd is a consultancy and solution provider specialising in information extraction, semantic annotation and search, linguistic engineering, sentiment analysis and opinion mining. We advise, evaluate and implement solutions based on one of the most widely used human language processing system in the world, GATE.

We have an extensive knowledge and experience of working with linked open data resources such as DBPedia, YAGO etc. to address complex issues such as identity resolution (named entity disambiguation) and relation extraction at large scale. Our expertise covers all aspects of documents life cycle, from web-wide crawling and parsing to categorization, semantic processing and indexing and developing effective search interfaces to allow semantic search over billions of documents. We also have deep understanding of turning massive volumes of unstructured social media data into meaningful insights such as entity (or feature) focused sentiments and opinions.

Our Consulting Domains

  • Plagiarism Detection
  • Patents and Citation Processing
  • Medial Literature/Clinical Data Processing
  • Social Media Analysis
  • Digital Forensic Analysis
  • Financial Markets
  • Product Reviews Analysis
  • Environmental Science
and more...

Experience Portfolio


A large scale digital forensic analysis system used for analyzing large media images (terabytes) to produce summaries and linked data visualizations in order to assist analysts in their digital media investigation.

Role: Lead technology consultant and the brain behind how semantics can help identify data that is relevant in the domain.


A small exploratory project with the British Library and HR Wallingford (funded by JISC) to experiment with LOD ontologies for environmental science.

Role: Lead technical developer responsible for providing semantic search facilities to allow environmental researchers to search successfully using LOD vocabularies.


Large-scale, Cross-lingual Trend Mining and Summarisation of Real-time Media Streams (funded by EC)

Role: Responsible for processing financial and political data from live streams such as twitter, news websites, blogs etc.

Responsible for developing an IE application to perform sentiment analysis on Tweets and News articles.


Large Knowledge Collider -- methods for retrieval and selection of data for reasoning to support carcinogenesis research (funded by EC).

Role: Responsible for mining literature knowledge to assist in predicting gene-disease associations in Genome Wide Association Studies (GWAS).

Our contributions enabled semantic search over the literature data and produced 1300 times faster results.


Developing an advanced indexing and search engine for Medical literature (funded by EC).

Role: Responsible for developing IE applications and annotating vast amount of medical data including radiology reports, Medline abstracts, Gene Home Reference articles, articles on internet related to health.

Use of Linked Life Data, UMLS, GATE and Metamap for IE, Nutch for crawling, Mimir for indexing of semantic content.

Developed a RESTFul XML API which turns Human Language queries into SPARQL and Mimir queries.


Semantic Annotations for Patent Processing (Industry funded project)

Role: Responsible for developing an IE system to process patents and extract information such as claims, citation references, section classification and author metadata.