SEARCH


inSitesEngine

Discover Value in Unstructured Text

inSitesEngine, RarePlay’s unique platform for distilling business intelligence from the web, utilizes innovative Natural Language Processing (NLP) and Machine Learning (ML) techniques to deliver critical business insight.

By reviewing and interpreting volumes of unstructured text including web pages, news articles, press releases, and publications inSitesEngine allows our clients to leverage the vast stores of buried knowledge in the internet (or private intranet) for key business decision-making.

Capitalize on the information available on the web and gain valuable business insight with RarePlay's platform for distilling unstructured online data into useful information.

inSitesEngine’s 12 Areas of Media Enhancement


Intelligent Fact Extraction (IFE)

inSitesEngine’s IFE reviews stores of unstructured digital media (internet/extranet files, blogs, chat room dialogues, online newspapers, advertisements, etc.), pulls out relevant information, and delivers insight in a translated, easily understandable format.

Key targets of interest of IFE include:
  • Data
  • Named-entity extraction (people, positions, companies, dates, locations, etc.)
  • Meta data
  • Trends
  • Themes
  • Consumer-generated media measurements
  • Sentiment
  • Demographics (age, gender, race, etc.)
  • Product/brand intelligence
  • Sentiment detection
  • Blog measurement
  • Trend identification
Key Concepts: Named-Entity Extraction, Information Retrieval, Business Intelligence, Knowledge Discovery, Neural Networks, Hidden Markov Model, Maximum Entropy

Interactive Prediction Systems (IPS)

Maximize your site’s efficiency with RarePlay and inSitesEngine’s IPS. Adapt your site to your clients – allowing you to more effectively communicate your messages to your audience by placing them where they are.

IPS equips your site with:
  • Adaptive site navigation (based on visitor paths)
  • Message routing (based on previous behavior)
Key Concepts: Software Agent

Decision Support Systems (DSS)

Employ inSitesEngine’s DSS to help your audience select the best course of action in situations where it is unclear where they need to navigate to. DSS guide users through logical decisions tress that direct them to information relevant to their query.

DSS systems include:
  • Product selectors
  • Plagiarism detectors
  • Expert systems
  • Self-service software
Key Concepts: Genetic Algorithms, Decision Trees

Content Recommendation Systems (CRS)

Implemented as a platform for discovery, inSitesEngine utilizes your audience’s defined preferences for a specific entity to locate and deliver new, relevant entities of interest. inSitesEngine detects and measures relevance levels based on patterns in the content.

CRS can be applied to numerable fields of interest, including:
  • Leisure
  • Music
  • Movies
  • Restaurants, etc.
  • Social
  • Dating
  • Activity partners
  • Professional
  • Job applicants
  • Prospective employers
Key Concepts: Collaborative Filtering, Computer-Supported Cooperative Work, Expert Locators, Expertise Finding, Expertise Location, Information Seeking, Recommendation Systems

Content Analysis Systems (CAS)

inSitesEngine’s CAS review and interpret available unstructured content to identify defined content variables. CAS provides tangible insight based on a collection of data;

Some of the key deliverables of CAS discovery are:
  • Brand messages
  • Reading levels
  • Target audience
  • Speech recognition
  • Translation
Key Concepts: Meta-Knowledge Discovery/Representation

Collaborative Filtering (CF)

inSitesEngine’s CF adds value to your audience’s user experience by delivering relevant options in tandem with the original search queries. Each new option presented is based on the similarity of preferences between all users who have initiated that particular query.

Key Concepts: Predictive Algorithms, Genetic Algorithms, Social Filtering

Intuitive Content Filtering (ICF)

ICF enhances the efficiency of your organizations online digital communications. Minimize the number of misdirected emails due to inaccurate labeling by users. ICF analyzes the content in each digital communication to ensure a more accurate distribution among the appropriate recipients.

ICF enhances an organization’s:
  • Spam filtering in emails
  • Corporate eMail routing
  • Customer service triage
  • Email classification
Key Concepts: Bayesian Algorithms, Neural Networks, Genetic Algorithms

Relationship Detection (RD)

inSitesEngine enables content clustering functionality. When integrated into applications RD identifies and defines the relationship between concepts and intuitively groups them based on the dynamics of their existing relationship.

Relationship Detection utilizes a number of user associations including:
  • Existing friendships or
  • Organization memberships
  • Work and education background
  • Geography and Language
  • Hobbies and points of interest
  • Etc.
Key Concepts: Content Clustering

Automatic Content Organization (ACO)

ACO streamlines large quantities of incoming digital media such as emails, articles, applications and estimates, etc. by analyzing and indexing them in appropriately defined categories. With ACO, users benefit from more efficient, simplified content navigation and review.

Key Concepts: Automatic Cataloging and Indexing, Neural Networks, Text Classification, Text Clustering

Content Summary Extraction (CSE)

CSE draws out and produces key concept summarizations from large unstructured messages so users can more efficiently review the relevance of the content. CSE minimizes time wasted on reviewing irrelevant materials and increases the efficiency of daily work flow.

Key Concepts: Bayesian Algorithms, Lexical Chains, Text Abstraction, Genetic Algorithm

Search

Facilitates users in locating the content they are looking for. Search powered by inSitesEngine delivers applicable search returns based on both keyword and concept driven queries. inSitesEngine search maximizes the accessibility of a organization’s information as well as the ability for users to locate and use it.

Key Concepts: Text Classification, Support Vector Machines, Semantic Indexing

Web Database (WebDB)

Web Database incorporates RegeXML, an xml-based search engine definition language, which enables users to effectively connect with and draw information from unstructured and partially structured web based information sources.

Key Concepts: Relational Algebra, Relational DBMS, Data Representation