ABSTRACT
Cyberspace has been recognized as a conducive environment for use of various hostile, direct, and indirect behavioural tactics to target individuals or groups. Denigration is one of the most frequently used cyberbullying ploys to actively damage, humiliate, and disparage the online reputation of target by sending, posting, or publishing cruel rumours, gossip, and untrue statements. Previous pertinent studies report detecting profane, vulgar, and offensive words primarily in the English language. This research puts forward a model to detect online denigration bullying in low-resource Hindi language using attention residual networks. The proposed model Hindi Denigrate Comment-Attention Residual Network (HDC-ARN) intends to uncover defamatory posts (denigrate comments) written in Hindi language which stake and vilify a person or an entity in public. Data with 942 denigrate comments and 1499 non-denigrate comments is scraped using certain hashtags from two recent trending events in India: Tablighi Jamaat spiked Covid-19 (April 2020, Event 1) and Sushant Singh Rajput Death (June 2020: Event 2). Only text-based features, that is, the actual content of the post, are considered. The pre-Trained word embedding for Hindi language from fastText is used. The model has three ResNet blocks with an attention layer that generates a post vector for a single input, which is passed through a sigmoid activation function to get the final output as either denigrate (positive class) or non-denigrate (negative class). An F-1 score of 0.642 is achieved on the dataset. © 2021 Association for Computing Machinery.
ABSTRACT
Maize is an important industrial crop where yield and quality enhancement both assume greater importance. Clean production technologies like conservation agriculture and integrated nutrient management hold the key to enhance productivity and quality besides improving soil health and environment. Hence, maize productivity and quality were assessed under a maize-wheat cropping system (MWCS) using four crop-establishment and tillage management practices [FBCT-FBCT (Flat bed-conventional tillage both in maize and wheat); RBCT-RBZT (Raised bed-CT in maize and raised bed-zero tillage in wheat); FBZT-FBZT (FBZT both in maize and wheat); PRBZT-PRBZT (Permanent raised bed-ZT both in maize and wheat], and five P-fertilization practices [P100 (100% soil applied-P); P50 + 2FSP (50% soil applied-P + 2 foliar-sprays of P through 2% DAP both in maize and wheat); P50 + PSB + AM-fungi; P50 + PSB + AMF + 2FSP; and P0 (100% NK with no-P)] in split-plot design replicated-thrice. Double zero-tilled PRBZT-PRBZT system significantly enhanced the maize grain, starch, protein and oil yield by 13.1-19% over conventional FBCT-FBCT. P50 + PSB + AMF + 2FSP, integrating soil applied-P, microbial-inoculants and foliar-P, had significantly higher grain, starch, protein and oil yield by 12.5-17.2% over P100 besides saving 34.7% fertilizer-P both in maize and on cropping-system basis. P50 + PSB + AMF + 2FSP again had significantly higher starch, lysine and tryptophan content by 4.6-10.4% over P100 due to sustained and synchronized P-bioavailability. Higher amylose content (24.1%) was observed in grains under P50 + PSB + AMF + 2FSP, a beneficial trait due to its lower glycemic-index highly required for diabetic patients, where current COVID-19 pandemic further necessitated the use of such dietary ingredients. Double zero-tilled PRBZT-PRBZT reported greater MUFA (oleic acid, 37.1%), MUFA: PUFA ratio and P/S index with 6.9% higher P/S index in corn-oil (an oil quality parameter highly required for heart-health) over RBCT-RBCT. MUFA, MUFA: PUFA ratio and P/S index were also higher under P50 + PSB + AMF + 2FSP; avowing the obvious role of foliar-P and microbial-inoculants in influencing maize fatty acid composition. Overall, double zero-tilled PRBZT-PRBZT with crop residue retention at 6 t/ha per year along with P50 + PSB + AMF + 2FSP while saving 34.7% fertilizer-P in MWCS, may prove beneficial in enhancing maize productivity and quality so as to reinforce the food and nutritional security besides boosting food, corn-oil and starch industry in south-Asia and collateral arid agro-ecologies across the globe.
ABSTRACT
During COVID-19, a large repository of relevant literature, termed as "CORD-19", was released by Allen Institute of AI. The repository being very large, and growing exponentially, concerned users are struggling to retrieve only required information from the documents. In this paper, we present a framework for generating focused summaries of journal articles. The summary is generated using a novel optimization mechanism to ensure that it definitely contains all essential scientific content. The parameters for summarization are drawn from the variables that are used for reporting scientific studies. We have evaluated our results on the CORD-19 dataset. The approach however is generic. © 2021 Polish Information Processing Society.
ABSTRACT
In this paper we present a scientific document retrieval system targeted at answering specific information needs for a diverse set of professionals. This became a crying need in the wake of the pandemic when researchers, clinicians, virologists, epidemiologists and health policy makers are regularly exploring a large collection of scientific literature related SARS-COV2 and COVID-19. The system facilitates exploration and easy comprehension of the results at multiple granularity through visual and textual summaries. The system uses novel NLP techniques to extract, aggregate and analyse textual information. The system also creates unique visual components to view key information of interest at individual and aggregate levels. The system is implemented as a web application. © 2021 Owner/Author.