Chandigarh University

About: Chandigarh University is a education organization based out in Mohali, India. It is known for research contribution in the topics: Materials science & Computer science. The organization has 1358 authors who have published 2104 publications receiving 10050 citations.

Analyzing DistilBERT for Sentiment Classification of Banking Financial News

Abstract: In this paper, the sentiment classification approaches are introduced in Indian banking, governmental and global news. The study assesses state-of-art deep contextual language representation, DistilBERT, and traditional context-independent system, TF-IDF, on multiclass (positive, negative, and neutral) sentiment classification news-events. The DistilBERT model is fine-tuned and fed into four supervised machine learning classifiers Random Forest, Decision Tree, Logistic Regression, and Linear SVC, and similarly with baseline TF-IDF. The findings indicate that DistilBERT can transfer basic semantic understanding to further domains and lead to greater accuracy than the baseline TF-IDF. The results also suggest that Random Forest with DistilBERT leads to higher accuracy than other ML classifiers. The Random Forest with DistilBERT achieves 78% accuracy, which is 7% more than with TF-IDF.

Experimental investigations on the performance and emissions characteristics of dual biodiesel blends on a varying compression ratio diesel engine

Abstract: The present work discusses the performance and emissions characterization of dual biodiesel sample blends on a varying compression ratio diesel engine. The dual biodiesel blends were obtained by blending two biodiesels (Mahua and Jatropha) in equal proportions volume (1:1, v/v) with mineral diesel. The sample blends were obtained on a ‘percentage by volume’ basis and named B10, B20, B30, and B40 (B10 was a blend of 5% each biodiesel with 90% mineral diesel and similarly for all other sample blends). All the experiments were performed at a constant engine speed of 1500 rpm, 50% loading conditions (2.6 kW), and varying compression ratios of 13.5:1, 14.5:1, 15.5:1, and 16.5:1. The results revealed that the sample blends had slightly higher brake power and mechanical efficiency with sample blends B10 to B40 had (0.15–1.58%) higher brake power and (1.07–12.42%) higher mechanical efficiency as compared to mineral diesel at a compression ratio of 16.5:1. The In-cylinder peak pressure and exhaust gas temperature were observed to be lower than mineral diesel for the sample blends B10 to B40 by 0.15–0.36 bar and 11.1–69.8 ℃, respectively. Also, the emissions of carbon monoxide and hydrocarbons were lower by 33–62%, respectively, for the sample with the highest blend percentage. However, the carbon dioxide emissions were found to be higher by 42.85% than mineral diesel. From the overall performance and characterization, it is concluded that B20 had optimum properties and blend percentage to be a better substitute fuel for mineral diesel among all the tested samples.

Iodine Catalyzed Oxidative Coupling of Diaminoazines and Amines for the Synthesis of 3,5-Disubstituted-1,2,4-Triazoles.

Abstract: A simple, convenient, transition metal-free one pot synthesis of 3,5-disubstituted-1,2,4-triazoles has been established. The innovation in this reaction is the use of easily available 1,1-diaminoazines as substrates. This method provides the products with wider substrate scope, at an expedited rate, and with relatively better yields in comparison to the reported methods. The reaction mechanism involves an initial intermolecular nucleophilic addition (facilitated by I2) followed by intramolecular nucleophilic cyclization.

An Efficient Data Delivery Scheme in WBAN to Deal with Shadow Effect due to Postural Mobility

Abstract: The body movement and change in posture exhibit high mobility in sensor nodes which causes shadowing in the Wireless Body Area Network (WBAN). Due to this, the connectivity between the nodes in WBAN is affected which further causes failure in data delivery. This article presents a MAC protocol in WBAN to deal with the problem of data delivery due to body movement and postural mobility. It uses an Improved Initial Centroid K-means clustering technique for classification of various human body postures followed by back propagation neural network as a classifier to recognize human body posture. This article proposes a posture aware dynamic data delivery (PA-DDD) protocol to deliver data dynamically. The PA-DDD protocol can be used under varying speed walking scenario. The simulation results show that it prolongs the network lifetime and is energy efficient.