Threatening language detection from Urdu data with deep sequential model.

The Urdu language is spoken and written on different social media platforms like Twitter, WhatsApp, Facebook, and YouTube. However, due to the lack of Urdu Language Processing (ULP) libraries, it is quite challenging to identify threats from textual and sequential data on the social media provided in Urdu. Therefore, it is required to preprocess the Urdu data as efficiently as English by creating different stemming and data cleaning libraries for Urdu data. Different lexical and machine learning-based techniques are introduced in the literature, but all of these are limited to the unavailability of online Urdu vocabulary. This research has introduced Urdu language vocabulary, including a stop words list and a stemming dictionary to preprocess Urdu data as efficiently as English. This reduced the input size of the Urdu language sentences and removed redundant and noisy information. Finally, a deep sequential model based on Long Short-Term Memory (LSTM) units is trained on the efficiently preprocessed, evaluated, and tested. Our proposed methodology resulted in good prediction performance, i.e., an accuracy of 82%, which is greater than the existing methods.

The impact of design on improved learning in virtual worlds: an experimental study.

Virtual worlds are the most advanced form of virtual environments, which offer one of the best platforms for serving various domains. They are, especially, well suited for education, to cope with the physical restrictions imposed due to COVID-19 outbreak, as they offer classroom experience to their users through immersion. They are online interactive spaces which are collaborative, persistent, coherent, and social in nature. Users immersed in these spaces are represented in the form of digital characters called, avatars. Virtual worlds offer advanced navigation methods such as flying and teleporting to facilitate quick learning. This paper analyses the use of a partial but carefully reconstructed cultural heritage site, developed in OpenSimulator framework, for learning both in terms of discourse and quantitative analysis. Discourse analysis compares the developed virtual world presence with traditional content provisioning methods in terms of a large set of well-known characteristics. Quantitative analysis, on the other hand, is based on data collected from users after conducting simple learning experiments. It revealed that the properties such as realism, friendliness, advanced navigation, and being detailed and social in nature greatly attracted user attention in learning. The learning was fast compared with traditional methods, however, it was a little hard for naive users to start exploring the content. Pre and post learning responses of users revealed that their knowledge level was significantly increased. Based on valuable suggestions, it is planned in future, to add intelligence to traditional agents, so they may help in an increased learning experience of users, based on the knowledge gained in earlier sessions.