Indian Institute of Technology Guwahati

About: Indian Institute of Technology Guwahati is a education organization based out in Guwahati, Assam, India. It is known for research contribution in the topics: Adsorption & Catalysis. The organization has 6933 authors who have published 17102 publications receiving 257351 citations.

Deep Learning for Detecting Cyberbullying Across Multiple Social Media Platforms

Abstract: Harassment by cyberbullies is a significant phenomenon on the social media. Existing works for cyberbullying detection have at least one of the following three bottlenecks. First, they target only one particular social media platform (SMP). Second, they address just one topic of cyberbullying. Third, they rely on carefully handcrafted features of the data. We show that deep learning based models can overcome all three bottlenecks. Knowledge learned by these models on one dataset can be transferred to other datasets. We performed extensive experiments using three real-world datasets: Formspring (\(\sim \)12k posts), Twitter (\(\sim \)16k posts), and Wikipedia(\(\sim \)100k posts). Our experiments provide several useful insights about cyberbullying detection. To the best of our knowledge, this is the first work that systematically analyzes cyberbullying detection on various topics across multiple SMPs using deep learning based models and transfer learning.

Supramolecular Peptide Amphiphile Vesicles through Host–Guest Complexation

Abstract: Single-tail peptide amphiphiles, have been explored as a new class of biomaterials in many fields including nanotechnology and tissue engineering. A typical peptide amphiphile molecule is linked through a covalent amide bond between a hydrophilic peptide sequence and a hydrophobic lipid of variable length. In an aqueous environment, these peptide amphiphiles undergo self-assembly into structures such as vesicles, both spherical and cylindrical micelles or nanotubes, and have been successfully applied in the biomedical sciences for biomaterial conjugation and as bioactive scaffolds for tissue engineering. Although covalent attachment of two components to form peptide amphiphiles has been extremely successful, the synthetic versatility and the ability to respond to external triggers remains limited. A supramolecular approach to form the peptide amphiphile by connecting two building blocks through a non-covalent interaction would represent a major advance, especially in designing stimuliresponsive systems capable of being targeted by specific triggers. Cucurbit[n]urils (CB[n]) are a family of macrocyclic hosts known to form inclusion complexes with selectivity and high binding affinity in aqueous media. One of the larger macrocycles in this family, CB[8], can be used as a “molecular handcuff” to join two molecules together in a non-covalent fashion, and has been applied to form biomaterials such as polymer–protein conjugation and protein dimerization. Additionally, CB[n] hosts have found great utility in “switch on/switch off” fluorescence assays by supramolecular complexation with various fluorescent guests. Pyrene and its derivatives have been widely used as fluorescence probes in a large number of complex systems, on account of their high fluorescence quantum yields, long excited state lifetimes and the ability to form excimers. Herein, we utilize a functional pyrene bearing an imidazolium group both as a fluorescence sensor and as a guest for CB[8] and linked it to a simple peptide sequence (1). Pyrene-functionalized peptide 1 is able to form the supramolecular peptide amphiphile complex 3 with viologen lipid 2 through CB[8] conjugation, as shown in Figure 1a. During the