Category Archives: Uncategorized

Gooru Algorithms (Deep-end tables)

This project is about implementation and validation of Gooru algorithms in the deep-end table. Deep end engineering is implemented in the form of a collection of “big tables” that contain materialized data relevant to several semantic operations. There are two big tables that come under deep-end big tables.

1) Activity Big Table: This table characterizes different elements of learning resources. These elements or columns include: Transcript, Summarization, Phrase Cloud, Competency, Relevance, Engagement, Efficacy, Instructional Practice and Rigor. This is part of curation of learning resources by using big data analytics.

It includes implementation of classification algorithm, relevance, engagement and efficacy scores. After computation of the scores for a subset of resources in datascope db, these scores will be computed for all the classified resources (3.4 million) from nucleaus db and will be populated for all the classified resources in the ABT of deep-end. Each of these columns are computed for a pair of learning resource and a competency.

2) Learning Big Table: This table characterizes a learner’s learning profile with respect to a subject. These elements or columns include performance, progress, proficiency, citizenship, authority, markers and goals.This is part of location of learner by using big data analytics.

It includes implementation of algorithms for computation of performance, progress, proficiency, authority and citizenship for learners and populate it into the LBT of deep-end.

Columns such as competency, transcripts, summarization, relevance, engagement, performance, progress, proficiency, authority and citizenship are implemented.

Implementation for computation of remaining columns is in progress.

Project Members:

1) Dr Aparna Lalingkar
2) Chaitali Diwan
3) Praseeda K
4) Udit Jindal
5) Ankit Gahalawat
6) Prasun Joshi
7) Palash Gupta
8) Urja Kothari
9) Rachna Shriwas
10) Shreyans Vora
11) Arjun Singh Kanwal

OSL Publications


Srinath Srinivasa. Aggregating Operational Knowledge in Community Settings. Proceedings of ODBASE 2012, Springer LNCS, Rome, Italy, Sep 2012.


Sanket Patil, Srinath Srinivasa. Theoretical Notes on Regular Graphs as applied to Optimal Network Design. Proceedings of the International Conference on Distributed Computing and Internet Technology (ICDCIT 2010), Bhubaneswar, India, February 2010.


Aditya Ramana Rachakonda, Srinath Srinivasa. Vector Based Ranking Techniques for finding Topical Anchors of a Context. Proceedings of COMAD 2009, Mysore, India, December 2009.

Saikat Mukherjee, Srinath Srinivasa, Krithi Ramamritham. On the Complexity of Multi-Query Optimization in Stream Grids. Proceedings of COMAD 2009, Mysore, India, December 2009.

Aditya Ramana Rachakonda, Srinath Srinivasa. Finding the Topical Anchors of a Context using Lexical Cooccurrence Data. Proc. of the 18th ACM Int’l Conf. on Information and Knowledge Management (CIKM’09), Hong Kong, China, Nov 2009.

Saikat Mukherjee, Srinath Srinivasa, Krithi Ramamritham. An Autonomous Agent Approach to Query Optimization in Stream Grids. Proceedings of the ACM Int’l Conference on Management of Digital Ecosystems (MEDES ’09), Lyon, France, October 2009.

Sanket Patil, Srinath Srinivasa, and Venkat Venkatasubramanian. Classes of Optimal Network Topologies under Multiple Efficiency and Robustness Constraints. Proc. of the IEEE Int’l Conference on Systems, Man and Cybernetics (SMC 2009), San Antonio, Texas, USA, October 2009, pp. 4940 – 4945


Karthik B R, Aditya Ramana Rachakonda, Srinath Srinivasa. Query Heartbeat: A Strange Property of Keyword Queries on the Web. Proc. of COMAD 2008, Mumbai, India, December 2008.


Saikat Mukherjee, Srinath Srinivasa, Saugat Mitra. WAND: A Robust P2P File System Tree Interface. International Conference on Distributed Computing and Internet Technology (ICDCIT 2007), Banglore, India.

Saikat Mukherjee, Srinath Srinivasa, Sanket Patil. Emergent (Re)Optimization for Stream Queries in Grids. Proceedings of the 2007 IEEE Congress on Evolutionary Computation (CEC 2007), IEEE Computer Society Press, September 2007.

Saikat Mukherjee, Srinath Srinivasa, Satish Chandra D. Validating for Liveness in Hidden Adversary Systems. Proceedings of the Third International Workshop on Foundations of Interactive Computing (FInCo 2007), Braga, Portugal, CWI Lecture Notes, March 2007.


Mistry Harjinder Singh, Srinath Srinivasa. Basis Graph: Combining Storage and Structure Index for Similarity Search in Graph Databases. Proc. of COMAD 2006, New Delhi, Tata McGraw-Hill, December 2006.

Siddhartha Reddy K, Srinath Srinivasa, Mandar R. Mutalikdesai. Measures of “Ignorance” on the Web. Proc. of COMAD 2006, New Delhi, Tata McGraw-Hill, December 2006.

Shibashis Guha, Srinath Srinivasa, Saikat Mukherjee, Ranajoy Malakar. LogicFence: A Framework for Enforcing Global Integrity Constraints at Runtime. Proceedings of IDEAS 2006, IEEE Computer Society Press, New Delhi, India, December 2006.

Aditya Ramana Rachakonda, Srinath Srinivasa. Incremental Aggregation of Latent Semantics Using a Graph-Based Energy Model. Proceedings of the International Conference on String Processing and Information Retrieval (SPIRE 2006), Glasgow, UK, Springer-Verlag, October 2006.

Mandar R. Mutalikdesai, Srinath Srinivasa. An Online Analytical Framework for Large Hypertext Collections. Proc. of VLDB Doctoral Consortium, Seoul Korea, ECWR Lecture Notes, September 2006.


Srinath Srinivasa, Sanket Patil. A Symmetric Localization Algorithm for MANETs based on Collapsing Coordinates. Proceedings of the International Conference on High Performance Computing (HiPC), Goa, India, Springer-Verlag, December 2005.

Pradeep S, Chitra Ramachandran, Srinath Srinivasa. Towards Autonomic Websites based on Learning Automata. Proceedings of WWW2005 (Special interest tracks and posters), Chiba, Japan, ACM Press, May 2005.

Anubhav Bhatia, Saikat Mukherjee, Saugat Mitra, Srinath Srinivasa. WAND: A Meta-Data Maintenance System over the Internet. Proceedings of WWW2005 (Special interest tracks and posters), Chiba, Japan, ACM Press, May 2005.

Srinath Srinivasa, Martin Meier, Mandar R. Mutalikdesai, P.S. Gopinath, K.A. Gowrishankar. LWI and Safari: A New Index Structure and Query Model for Graph Databases. Proceedings of COMAD 2005, Goa, India, Tata McGraw-Hill, January 2005.


Dina Goldin, Srinath Srinivasa, Vijaya Srikanti. Active Databases as Information Systems. Proceedings of the International Database Engineering and Applications Symposium (IDEAS ’04), Coimbra, Portugal, Springer-Verlag, July 2004.


Srinath Srinivasa, Sujit Kumar. A Platform Based on the Multi-Dimensional Data Model for Analysis of Bio-Molecular Structures. Proceedings of VLDB 2003. Berlin, Germany, September 2003.

Sujit Kumar, Srinath Srinivasa. A Database for Storage and Fast Retrieval of Structures: A Demonstration. Proceedings of ICDE 2003, Bangalore, India, IEEE Computer Society Press, March 2003.


Srinath Srinivasa, Sumit Acharya, Himanshu Agrawal, Rajat Khare. Vectorization of Structure for Indexing Graph Databases. Proceedings of the IASTED International Conference on Information Systems and Databases (ISDB 2002), Tokyo, Japan, September 2002.

Talk on “Bridging big data and qualitative methods in the social sciences”

Date: Jan 3rd 2018

Time: 2:15 PM

Location: TBD

Abstract: With the rise of social media, a vast amount of new primary research material has become available to social scientists, but the sheer volume and variety of this make it difficult to access through the traditional approaches: close reading and nuanced interpretations of manual qualitative coding and analysis. This work sets out to bridge the gap by developing semi-automated replacements for manual coding through a mixture of crowdsourcing and machine learning, seeded by the development of a careful manual coding scheme from a small sample of data. To show the promise of this approach, we attempt to create a nuanced categorisation of responses on Twitter to several cases of extreme circumstances.


Bio of speaker:

Dima is a Senior Data Scientist at Skyscanner where his focus is on developing and optimizing the Skyscanner’s travel search engine. Prior to Skyscanner, Dima was with King’s College London where he worked on analysis of BBC iPlayer (a joint project with BBC) and various social media websites (Twitter, Pinterest, Foursquare, etc.). He contributes to the data mining (KDD, WWW, etc.) and computer networks communities (Infocom, ComMag, etc.) and have his works featured by New ScientistBBC News and other media outlets. Dima has also co-founded and was a former CEO of More information –


Gooru is a learning environment modeled as a social machine. It acts as a learning navigator. Any learning experience is in the form of a navigation across the “learning map” using “competencies”. Competencies are made up of two dimensions; namely pedagogic depth and topic. In order to learn a particular topic or to reach a target competency, Gooru provides the learner with an optimal learning route. The skyline of a learner is the set of the highest achieved competencies in every topic. Every time the learner learns a new topic or achieves a new competency, his skyline is updated. This gives the learner his position on the learning map. The learner skyline is used to compute the route the learner has to take to reach a target competency.

  • Project Members


Gooru honors the human right to education by creating technology that enables educators and researchers to “open-source” effective practices and content to improve learning outcomes for all.

The Gooru model of learning is a Computer Aided Instruction (CAI) environment that aims to provide effective, individualized learning experiences at scale, by utilizing both adaptive tutoring as well as community interactions. User interaction is modeled around a “Navigation” paradigm, where the subject matter is represented as learning activities scattered across a logical space called the “learning map.” Lessons are modeled as learning pathways in this logical space. The logical space metaphor enables the student to obtain different levels of overview of the subject matter by perusing the landscape, and understand semantic relatedness of topics based on their visual proximity. The learning map is also a social space, where students interact with other learners (other students, human tutors and teachers), rather than only an algorithmic agent.

The underlying paradigm of such a system is called a social machine. The social machine approach is characteristically different from both scalable classroom models like MOOCs and personalized learning environments like Intelligent Tutoring Systems

Sub Projects

Open City

Open City projects aims at large scale access control management. Large amount of data is generated by IOT devices installed in the city but its limited usage and difficult accessibility has lead to under utilization of resources and huge expenditure for e and the government. If we could create a system where such sensitive data is uploaded and the owner of the data could decide who should be provided access then  it would be of great help. We could get real time update of the traffic with the cameras installed outside buildings and we could manage the traffic better. We can have automated surveillance mechanisms built.

There is great advantage of building such a system, but there are equally large challenges involved. Misuse of open ended data could lead to people losing trust from the system.

Open City aims at building a system that would share relevant data based on some events and triggers and only to types of people defined by the owner of the data.