Title: Senior Researcher and Manager, Information and Analytics Department
Affiliation: IBM Research - India
Contact Details: email@example.com
Sameep Mehta is a Senior Researcher and Manager at IBM Research - India. He received his Ph.D. in Data Mining and Visualization from Ohio State University. His current research interests are Data Mining, Text Mining, Machine Learning, Big Data Technologies, Social Data Analytics and Knowledge Graph. He has published extensively in top conferences in Data Mining, Services and Visualization. He is a regular speaker at conferences and is PC chair for Big Data Analytics Conference 2014. He also serves as Adjunct Faculty at IIIT-Delhi in the area of Data Analytics.
Title of Talk 1: Introduction to Knowledge Graph Stores with Applications
Synopsis: This talk will give an introduction to popular Graph Stores like Titan, Neo4J, etc. We will motivate the applications for which graphs present a natural modeling choices. We will discuss popular triplet stores like RDFs and how they compare and contrast with graph stores. The talk will argue on the scalability aspect of the data and show how the volume of data is handled by graph stores at back end. The attendees will be exposed to common graph querying language and search capabilities including elastic search. The talk will focus on one end-to-end scenario to show different steps like data preparation, data ingestion and accessing the data through APIs. We will conclude the data to by showing a demo of application built on graphs.
Title of Talk 2: Data Fusion Enabled Contextual and Personalized Insight Mining
Synopsis: In this talk, we will provide a motivation for data fusion to generate actionable insights and to support analytics which could not have been supported in siloed data sources. We will propose the use of graph stores for hosting the common fused data. Graphs are very natural choice for storing variety of data and relationship between the data. We will show how missing/partial data problem can be handled by using graphs and associated inferencing algorithms. A typical life cycle of multiple data sources to graph will be demonstrated through a case study including the applications that be built on top of it. We will argue how the value of the siloed data sources increase considerably after fusing. The talk will be concluded by going over some open research issues in this area.