Innovative Vision Requires Innovative Solutions

The demand for information retrieval strategies is at its highest. As areal densities continue to push the superparamagnetic limit, storage capacities continue to increase. More than ever, companies are finding themselves drowning in a sea of information.

The R&D group at ESR is primarily focused on building the next generation search experience. This involves innovative add-on features to existing information retrieval systems as well as an active role in the study and development of search engine technology.

Content Aggregation Strategies
One of the most fundamental tasks of building IR systems is content aggregation. Customer data is often comprised of various formats that span numerous storage mechanisms in multiple physical locations. The R&D group at ESR develops innovative and scalable solutions that allow the customer to aggregate this content into a single unified searchable index. Some current areas of research and development include:

  • Enterprise Scaling with Asynchronous Network IO
  • Enterprise Scaling with the MapReduce Algorithm
  • Language Bindings for High Level Languages
  • Document Filtering and Content Cleansing

Semantic Search Enhancements
Modern commercial search engines provide the ability to search for information based on semantic meaning. We have worked extensively with some of the leading search vendors to enhance this experience by utilizing natural language processing techniques as a way to add semantic meaning to unstructured text. In particular, we have experience in:

  • Named Entity Recognition
  • Anaphora Resolution
  • Part of Speech Tagging
  • Natural Language Processing
  • Sentiment Analysis

Collective Intelligence
 The Internet has given way to new forms of communication technology and with this comes the advent of collective intelligence. The study of collective intelligence has been around for decades but new technology breaths new life into the possibilities. The core of this work is focused on:

  • Decision Tree Modeling
  • Naïve Classifiers
  • Advanced Classification using Kernel Methods & Support Vector Machines
  • Machine Learning

Rich Internet Applications
The buzz of Enterprise 2.0 is paving the way for a new wave of business communication tools that enable contextual, agile and simplified information exchange and retrieval. Utilizing emerging technologies, ESR is researching ways to help customers build the next generation of business tools. Some of these emerging technologies include:

  • JavaServer Faces (JSF)
  • JSF Custom Components & Data Providers
  • Asynchronous Javascript and XML (Ajax)
  • Dojo
  • XML Rich Client Technology
  • Python Web Frameworks such as Zope, Django, and Turbo Gears
  • Web Server Gateway Interface (WSGI)

Information Retrieval Systems
At ESR, search is our business so we make it point to be actively involved in the science behind information retrieval. From theory to praxis, the R&D group maintains an intimate knowledge of the various techniques used in search engine design including:

  • Traditional Boolean Retrieval Model
  • Ranked Retrieval Model
  • Assessing Relevancy Through Precision & Recall Metrics
  • Vector Space Search Engine Theory
  • Query Optimization
  • Tokenization and Linguistic Analyzers

Let ESR help your vision become a reality.