Spring

In libraries, books are classified into some categories according to their topics in order to be found easily. However, due to the great expansion of Internet, there are enormous number of texts online and classifying them by hand is not possible. So, we need text classification algorithms. In literature, there are many types of text classification algorithms but they do not work well in skewed datasets. Skewed datasets are the datasets that have huge difference between number of documents of its classes.

For the last two decades Chess has been a trending topic among Artificial Intelligence researchers. Since chess is considered to be an act of pure intelligence and grandmaster have been considered as possessing the most complex intelligence in the world, researchers in the earlier times of the AI focused on creating strong chess playing programs. Claude Shannon and Alan Turing made serious contributions for building the foundations of chess engines.

Deep neural networks have had huge success in recent years. However training them efficiently is still a research area. In this project we investigated a type of neural network, autoencoder, and its variations. Autoencoder is an ANN that tries to reconstruct its input as output. While doing this, it learns features of data as a byproduct. It is also used as a pre-training tool for deep neural networks.

Annotation is way of associating information with resources. Recently, World Wide Web Consortium (W3C) published annotation data model[23 February 2017]. Purpose of this project was to find a use case for
annotation data model. The use case was Crowdmap. CrowdMap is an application that allows users to collaboratively add meta information or relations between subjects. It is reminiscent of the detective walls with pictures of suspects and maps that are annotated via detectives.