Data and Big Data Analytics have recently emerged as major areas of focus. Reveal is one of the leading providers of skilled professionals with this critical skill set. Our database managers are top notch professionals that deliver cutting edge solutions to our customers which provides them with the tools for improved decision-making and increased efficiency.

Reveal utilizes Business Intelligence (BI) as a tool for the transformation of raw data into meaningful and useful information for analysis purposes. BI technologies are capable of handling large amounts of unstructured data to help identify, develop, and otherwise create new strategic business opportunities.

Common functions of business intelligence technologies are reporting, online analytical processing, analytics, data mining, process mining, complex event processing, business performance management, benchmarking, text mining, predictive analytics, and prescriptive analytics. We provide this intelligence for the easy interpretation of large volumes of data. Identifying new opportunities and implementing an effective strategy based on insights can provide businesses with a competitive market advantage and long-term stability.

Data Quality Approach

Scoring of Algorithms

• Supervised algorithms are scored – algorithm scores denote the percentage of records where the supervised algorithm is in agreement with the training data
• Only algorithms that score greater than 95% are selected for analysis, which enhances the confidence in the results

Statistical Validation

• Assign threshold fitness value (-1.0) to records to separate” inlier” (FV> threshold) from “outlier” records
• Label each row as “inliers” or “outliers” and feed these labels during training of supervised algorithms
• Calculate unsupervised FV for each row and the distance of the row from the decision surface of each of the four unsupervised learning algorithms
• Experimentation demonstrates that “outlierness” correlates to more negative FV and “inliers” have FV closer to +1.0. Outliers show a strong disagreement between unsupervised and supervised FVs.
• Ran the analytics program twice with different FV ranges (-20, -5 & -5, +1) to identify extreme outliers and those records on the cusp of being outlier/inlier.

Data Quality Validation

Develop an AI based expert system for a US Agency to analyze agency data, deduce new knowledge and provide tactical support to field personnel

• Natural Language Processing – leveraging RAC capabilities to ingest and analyze data from various sources (social media, newsfeeds, reports, video and audio communication, etc.)
• Apply AML techniques to discover relationships between entities and subjects that are not apparent, define strengths of the relationships through graphical representation
• RAC provides a battle order (org chart/deck of cards) based on the analysis that the agency is using in the field
• Identify abnormal/outlier behavior or unusual relationships between actors and report those to the agency for deeper analysis and on-field validation
• Provide context specific recommendations which allows the user to create intelligent Q&A sessions
• Recommendations/findings are input into the agency’s analytic support engine where users can evaluate and take further action as necessary