3^rd Workshop on Machine Learning & Data Analytics

Topics to be covered

Machine Learning Algorithms

• Feature Engineering- Non-linear dimensionality reduction

• Kernel based learning

• Probabilistic learning

• Transfer learning

• Fairness, Accountability and Transparency

Data Mining and Analytics

• Social Network Mining: The data generated by social networks like Instagram provide valuable information about the behavior, tastes, preferences, and other characteristics of users. Transformation of this data into relevant information is a challenge and had been proven to be very effective to generate value. The workshop will help participants in understanding some existing solutions and will elaborate some of the existing challenges in the area.

• Multi-relational Data Mining: In order to extract important information or knowledge, it is required to apply Data Mining algorithms on this relational database but most of these algorithms works only on single table. To overcome this challenge generation of a single table may result in to loss of important information, like relation between two different tuples. The workshop will help participants in understanding MRDM approaches and there different application domains like, counter terrorism, viral marketing, social networks, computational biology, ubiquitous computing etc.

• Mining on Big Data: Data with multiple features and attributes which makes data analysis complex is called high dimensional data (HDD). Problems with high dimension data are like, difficult visualization, intractable subspace enumeration, less precise distance measure, feature correlation etc. Clustering of such high dimension datasets is a challenge and one section of the workshop is dedicated for elaboration of approaches which can be used to handle high dimension data and perform clustering over such data.

• High-Dimensional Data Clustering: Continuous increase in amount of data in all science and engineering domains demanding scalable approaches for extraction of information from this huge amount of data. This large amount of data taking form of BigData often stored in a distributed environment. So, there is urgent need of developing effective mining processes which can accommodate this distribution of data.

Data Security and Privacy

• Data Analytics privacy: Privacy information may be leaked while analyzing the data for statistics or other requirements. It is necessary to protect the privacy of information at the same time near accuracy. There are cryptographic and non-cryptographic techniques available to protect the privacy with a tradeoff of accuracy. It is an important topic since data analytics plays major role in health, business and astronomical data.

• Security and Privacy in remote storage service: Data stored at the third party servers and provide service to clients will make the service efficient. However, data owner will face the security and privacy issue since data is not in their control. Hence a proper security mechanism should be applied. In this workshop, we will be discussing some of the mechanisms.

• Blockchain: It is an emerging platform to implement the decentralized applications. It will be used as a solution for the untrusted environment. It is an effiecient solution for key distribution, authentication, registry maintenance, etc.