http://repository.iiitd.edu.in/xmlui/handle/123456789/888 Fri, 10 Apr 2026 20:02:30 GMT 2026-04-10T20:02:30Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1188 Characterization of novel short proteins in the human microbiome Mrinal; Gautam, Prutyay; Ray, Arjun (Advisor) The project is based on applying Machine Learning tools and di erent analysis techniques to identify which protein features in metagenomes sampled from Human and Non-Human (other mammal species) microbiomes help them impart species speci city. This task would be followed by experimental validation of these identi ed features. Tue, 01 Dec 2020 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1188 2020-12-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1130 Analysis and modelling of disfluency and as units for automatic scoring of L2 speakers Mohan, Gandharv; Sharma, Akash; Shah, Rajiv Ratn (Advisor) In this project, we attempt to understand how dis uencies in speech can be used in providing feedback for automated oral pro ciency scores in the context of spontaneous speech for L2 learn- ers of English. Our aim is to model and present a corpus of spontaneous speech of non native speakers of English consisting of segment labelled annotations of dis uencies. This corpus will be further used to develop dis uency detection models for automated scoring systems of oral pro ciency. In this report, we discuss some crucial points that we covered from our extensive literature review on dis uency from linguistics point of view as well as computational point of view. Then, through these discussions we prepare an annotation scheme for annotating dis u- encies for our dataset targeted speci cally for our purpose. We also do a qualitative analysis of the dataset to nd out more interesting features which could be helpful in shaping our future work. Tue, 01 Dec 2020 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1130 2020-12-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1016 Analysing social media for COVID positive cases Singhal, Sonali; Pandey, Tanisha; Kumaraguru, Ponnurangam (Advisor); Shah, Rajiv Ratn (Advisor) The COVID19 pandemic, a public health emergency which broke out globally in December 2019 and is still going on, has put us all into an unprecedented situation where there are strict limitations on physical social interaction and everyone is confined to their homes. This has increased importance of online support from social media in our lives. With social media becoming a major source of help, consolation, medical information, and a platform for sharing our own experiences. This study is a deep dive into the various roles that social media has played during this pandemic. We try to answer social questions such as what is being talked about in the online COVID communities, how testing positive changes people's online behaviour, what are the most prominent symptoms during COVID and what is the effect of giving and receiving support from the community. Finally, we classify the dataset in terms of COVID phases, support and user behavior and analyse how support affects people in various phases. Our results lead to a better understanding of how users are getting affected in this unusual situation and study how online social communities can support each other in such trying times. Fri, 01 May 2020 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1016 2020-05-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1007 Emoji analysis: a study on the use of skin tone modifiers on social media Bansal, Tanmay; Kumaraguru, Ponnurangam (Advisor) In 2015, Unicode introduced five different skin tone modifiers, that could be applied to the existing emojis that denoted humans. In this project, we analyze how people pose certain personalities using emoji skin colour modifiers under different situations. A simple frequency-based analysis was carried out on the usage of emoji and their skin tone modifiers. We also recorded that how the skin tone modifier changes when user uses hash tags or mentions other Twitter users. Our analysis reveals that with high probability a user will always use the same skin tone modifier which was used in the previous tweet. Dark skinned users are most likely to deviate from their base tone. Also, we found out that users mostly switch skin tone modifier when they either post a tweet mentioning exactly 1 person, or a plain text tweet with no hash tags or mention. Thus, it shows that people reveal multiple identities through tone modifiable emojis. Tue, 01 Dec 2020 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1007 2020-12-01T00:00:00Z