Browsing by Author "Singh, Satwinder"

Now showing 1 - 20 of 26

Advances and challenges in thyroid cancer: The interplay of genetic modulators, targeted therapies, and AI-driven approaches
(Elsevier Inc., 2023-09-20T00:00:00) Bhattacharya, Srinjan; Mahato, Rahul Kumar; Singh, Satwinder; Bhatti, Gurjit Kaur; Mastana, Sarabjit Singh; Bhatti, Jasvinder Singh
Thyroid cancer continues to exhibit a rising incidence globally, predominantly affecting women. Despite stable mortality rates, the unique characteristics of thyroid carcinoma warrant a distinct approach. Differentiated thyroid cancer, comprising most cases, is effectively managed through standard treatments such as thyroidectomy and radioiodine therapy. However, rarer variants, including anaplastic thyroid carcinoma, necessitate specialized interventions, often employing targeted therapies. Although these drugs focus on symptom management, they are not curative. This review delves into the fundamental modulators of thyroid cancers, encompassing genetic, epigenetic, and non-coding RNA factors while exploring their intricate interplay and influence. Epigenetic modifications directly affect the expression of causal genes, while long non-coding RNAs impact the function and expression of micro-RNAs, culminating in tumorigenesis. Additionally, this article provides a concise overview of the advantages and disadvantages associated with pharmacological and non-pharmacological therapeutic interventions in thyroid cancer. Furthermore, with technological advancements, integrating modern software and computing into healthcare and medical practices has become increasingly prevalent. Artificial intelligence and machine learning techniques hold the potential to predict treatment outcomes, analyze data, and develop personalized therapeutic approaches catering to patient specificity. In thyroid cancer, cutting-edge machine learning and deep learning technologies analyze factors such as ultrasonography results for tumor textures and biopsy samples from fine needle aspirations, paving the way for a more accurate and effective therapeutic landscape in the near future. � 2023 The Author(s)
Analyze Dark Web and Security Threats
(Springer Science and Business Media Deutschland GmbH, 2023-05-03T00:00:00) Ansh, Samar; Singh, Satwinder
The deepest area of data storage where data mining and data management are not possible without the Tor (network) Policy is known as the dark web. The dark web is a paradise for government and private sponsored cybercriminals. In another word, the dark web is known as the underworld of the Internet used for sponsored and organized cybercrime. Tor network at the entry relay/guard user source IP replaced with local IP (i.e., 10.0.2.15) by default and every user machine ID (IP) recognize as local IP (10.0.2.15). A single source IP allocated for each user without collision makes the user an anomaly or invisible over the Internet. Tor browser works similar to VPN by default as a function to hide the source IP, but the advantage is Tor network�s volunteer devices are used as a tunnel to establish communication and offer freedom from surveillance of user activity. Tor browser offers a circuit (IP Route) for user activity, where the circuit allows available Tor IP at the exit relay for the user. The dark web uses the same IP at entry relay around the world, but at exit relay, IP is different and available based on country. In a dark web network, data transfer as an encapsulation of packet/massage is placed after three-layer of different encryption. Proposed six different machine-learning classifiers (Logistic Regression, Random Forest, Gradient Boosting, Ada Boosts, K-Nearest Neighbors, Decision Tree) used to the optimal solution and proceed to analyze security threats perform in the Dark web based on the communication protocol and user activity as data flow and active state. � 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
Bug Classification Depend Upon Refactoring Area of Code
(Springer, 2023-01-05T00:00:00) Singh, Satwinder; Jalal, Maddassar; Kaur, Sharanpreet
Due to rapid development in the software industry, various software is being developed, which compromises the software quality. As long as time passes, the software which compromises with the quality starts showing bugs which adversely affect the working of the software system. Sometimes software undergo repeated addition of functionality, so various classes and functions in software systems become bulky and cause smells in code. Code smells also degrade the software quality, increasing the software system's maintenance cost. So these bad smells should be appropriately detected and eliminated from the software system well on time. With the identification of smells in code, one could easily map the refactoring area of code to improve the code. Improving the code will help in preventive maintenance to identify the bugs quickly. To solve this problem, the current study will focus on the five types of code smells: Data Class, God Class, Feature Envy, Refuse Parent Request and Brain method for identification of bugs in the code. The data set used for this study includes the smell extraction and identification from two versions of Eclipse, i.e. Eclipse 3.6 and Eclipse 3.7, which are renowned open soured industry size software systems. Later on, supervised machine learning classifiers j48, Random Forest, and Naive Bayes are used to identify the bugs in code at the class level. The results show that j48 performed well and provides high accuracy for the identification of bugs with the help of bad smells. � 2023, The Institution of Engineers (India).
Classification of defective modules using object-oriented metrics
(Inderscience Enterprises Ltd., 2017) Singh, Satwinder; Singla, Rozy
Software defect in today's era is crucial in the field of software engineering. Most of the organisations use various techniques to predict defects in their products before they are delivered. Defect prediction techniques help the organisations to use their resources effectively which results in lower cost and time requirements. There are various techniques that are used for predicting defects in software before it has to be delivered, e.g., clustering, neural networks, support vector machine (SVM). In this paper two defect prediction techniques: K-means clustering and multi-layer perceptron model (MLP) are compared. Both the techniques are implemented on different platforms. K-means clustering is implemented using WEKA tool and MLP is implemented using SPSS. The results are compared to find which algorithm produces better results. In this paper object-oriented metrics are used for predicting defects in the software. Copyright ? 2017 Inderscience Enterprises Ltd.
Code clone detection and analysis using software metrics and neural network: A Literature Review
(Eighth Sense Research Group, 2015) Kumar, Balwinder; Singh, Satwinder
Code clones are the duplicated code which degrade the software quality and hence increase the maintenance cost. Detection of clones in a large software system is very tedious tasks but it is necessary to improve the design, structure and quality of the software products. Object oriented metrics like DIT, NOC, WMC, LCOM, Cyclomatic complexity and various types of methods and variables are the good indicator of code clone. Artificial neural network has immense detection and prediction capability. In this paper, various types of metric based clone detection approach and techniques are discussed. From the discussion it is concluded that clone detection using software metrics and artificial neural network is the best technique of code clone detection, analysis and clone prediction
Comparative performance of fault-prone prediction classes with k-means clustering and MLP
(Association for Computing Machinery, 2016) Singh, Satwinder; Singla, Rozy
Software defect in today's era is most important in the field of software engineering. Most of the organizations used various techniques to predict defects in their products before they are delivered. Defect prediction techniques help the organizations to use their resources effectively which results in lower cost and time requirements. There are various techniques that are used for predicting defects in software before it has to be delivered. For example clustering, neural networks, support vector machine (SVM) etc. In this paper two defect prediction techniques:-K-means Clustering and Multilayer Perceptron model (MLP), are compared. Both the techniques are implemented on different platforms. K-means clustering is implemented using WEKA tool and MLP is implemented using SPSS. The results are compared to find which algorithm produces better results. In this paper Object-Oriented metrics are used for predicting defects in the software. ? 2016 ACM.
Comparison and analysis of logistic regression, Na�ve Bayes and KNN machine learning algorithms for credit card fraud detection
(Springer Science and Business Media B.V., 2020-02-15T00:00:00) Itoo, Fayaz; Meenakshi; Singh, Satwinder
Financial fraud is a threat which is increasing on a greater pace and has a very bad impact over the economy, collaborative institutions and administration. Credit card transactions are increasing faster because of the advancement in internet technology which leads to high dependence over internet. With the up-gradation of technology and increase in usage of credit cards, fraud rates become challenge for economy. With inclusion of new security features in credit card transactions the fraudsters are also developing new patterns or loopholes to chase the transactions. As a result of which behavior of frauds and normal transactions change constantly. Also the problem with the credit card data is that it is highly skewed which leads to inefficient prediction of fraudulent transactions. In order to achieve the better result, imbalanced or skewed data is pre-processed with the re-sampling (over-sampling or under sampling) technique for better results. The three different proportions of datasets were used in this study and random under-sampling technique was used for skewed dataset. This work uses the three machine learning algorithms namely: logistic regression, Na�ve Bayes and K-nearest neighbour. The performance of these algorithms is recorded with their comparative analysis. The work is implemented in python and the performance of the algorithms is measured based on accuracy, sensitivity, specificity, precision, F-measure and area under curve. On the basis these measurements logistic regression based model for prediction of fraudulent was found to be a better in comparison to other prediction models developed from Na�ve Bayes and K-nearest neighbour. Better results are also seen by applying under sampling techniques over the data before developing the prediction model. � 2020, Bharati Vidyapeeth's Institute of Computer Applications and Management.
Comparison of Public and Critics Opinion About the Taliban Government Over Afghanistan Through Sentiment Analysis
(Springer Science and Business Media Deutschland GmbH, 2023-05-03T00:00:00) Reza, Md Majid; Singh, Satwinder; Kundra, Harish; Reza, Md Rashid
The usage of social media has increased exponentially these days. People worldwide are sharing their opinions on different platforms such as Twitter, personal blogs, Facebook, and other similar platforms. Twitter has grown in popularity as a platform for people to express their thoughts and opinions on many different topics. The data from Twitter about the Taliban has been examined in this research work, and various machine learning algorithms have been applied including SVM, LR, and random forest. Text sentiments have been captured via TextBlob. Among the machine learning models applied, SVM outperformed all other models and achieved an accuracy score of around 94% on the tweet dataset and logistic regression outperformed other models with an accuracy score of 83% on the news article dataset. � 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
Defect prediction model of static code features for cross-company and cross-project software
(Springer Science and Business Media B.V., 2018-12-06T00:00:00) Singh, Satwinder; Singla, Rozy
Software project metrics are seen needless in software industries but they are useful when some unacceptable situations come in the project (Satapathy et al., Proceedings of the 48th annual convention of CSI, vol 2, 2013). Mainly the focus of various defect prediction studies is to build prediction models using the regional data available within the company. So companies maintain a data repository where data of their past projects can be stored which can be used for defect prediction in the future. However, many companies do not follow this practice. In software engineering, the crucial task is Defect prediction. In this paper, a binary defect prediction model was built and examined if there is any conclusion or not. This paper presents the assets of cross-company and within-company data against software defect prediction. Neural network approach has been used to prepare the model for defect prediction. Further, this paper compares the results of with-in and cross-company defect prediction models. To analyse the results for with-in company two versions of Firefox (i.e. 2.0 and 3.0) were considered; for cross project one version of Mozila Sea Monkey (1.0.1); for cross-company validation one version of LICQ were considered. Main focus of the study is to analyse the behavior or role of software metrics for acceptable level of defect prediction. � 2018, Bharati Vidyapeeth's Institute of Computer Applications and Management.
Dual aromatase-steroid sulfatase inhibitors (DASI's) for the treatment of breast cancer: a structure guided ligand based designing approach
(Taylor and Francis Ltd., 2022-12-13T00:00:00) Singh, Yogesh; Jaswal, Shalini; Singh, Satwinder; Verma, Sant Kumar; Thareja, Suresh
Dual aromatase-steroid sulfatase inhibitors (DASIs) lead to significant deprivation of estrogen levels as compared to a single target inhibition and thereby exhibited an additive or synergistic effect in the treatment of hormone-dependent breast cancer (HDBC). Triazole-bearing DASI�s having structural features of clinically available aromatase inhibitors are identified as lead structures for optimization as DASI�s. To identify the spatial fingerprints of target-specific triazole as DASI�s, we have performed molecular docking assisted Gaussian field-based comparative 3D-QSAR studies on a dataset with dual aromatase-STS inhibitory activities. Separate contours were generated for both aromatase and steroid sulphates showing respective pharmacophoric structural requirements for optimal activity. These developed 3D-QSAR models also showed good statistical measures with the excellent predictive ability with PLS-generated validation constraints. Comparative steric, electrostatic, hydrophobic, HBA, and HBD features were elucidated using respective contour maps for selective target-specific favourable activity. Furthermore, the molecular docking was used for elucidating the mode of binding as DASI�s along with the MD simulation of 100 ns revealed that all the protease-ligand docked complexes are overall stable as compared to reference ligand (inhibitor ASD or Irosustat) complex. Further, the MM-GBSA study revealed that compound 24 binds to aromatase as well as STS active site with relatively lower binding energy than reference complex, respectively. A comparative study of these developed multitargeted QSAR models along with molecular docking and dynamics study can be employed for the optimization of drug candidates as DASI�s. Communicated by Ramaswamy H. Sarma. � 2022 Informa UK Limited, trading as Taylor & Francis Group.
Hate Speech and Offensive Language Detection in Twitter Data Using Machine Learning Classifiers
(Springer Science and Business Media Deutschland GmbH, 2023-05-03T00:00:00) Shah, Seyed Muzaffar Ahmad; Singh, Satwinder
Social media is rapidly growing in popularity and has its advantages and disadvantages. Users posting their daily updates and opinions on social media may inadvertently hurt the feelings of others. Detecting hate speech and harmful information on social media is critical these days, lest it led to calamity. In this research, machine learning classifiers such as Na�ve Bayes, support vector machines, logistic regression, and pre-trained models BERT and RoBERTa, developed by Google and Facebook, respectively, are used to detect hate speech and offensive content from Twitter data on a newly created dataset that included tweets and articles/blogs. The sentiments were obtained using the VADER sentiment analyzer. The results depicted that the pre-trained classifiers outperformed the machine learning classifiers utilized in this study. An accuracy score of 96% and 93% was scored by BERT and RoBERTa, respectively, on the tweet dataset, whereas on a dataset of articles/blogs, accuracy of 97% and 98%, respectively, was achieved by both the classifiers outperforming other classifiers used in this work. Further, it can also be depicted that neutral content is shared more in articles/blogs, hate content is mostly shared equally in both the tweets and article/blogs, whereas offensive content is shared higher in tweets than articles/blogs. � 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
Improving the Quality of Open Source Software
(wiley, 2023-02-08T00:00:00) Kaur, Sharanpreet; Singh, Satwinder
This study aims at development of generating metrics based code smells prediction to improve the software quality assurance by working at preventive maintenance level. In order to do so, Refactoring is the best solution for identification of smelly areas in the code to reveal the portions which demands patching. It not only increases the life of code but eventually increases the quality of software in long run, where versions of a software are launched one after the other. The empirical model development considered Deep learning based neural network technique for establishing the association between code smells and metrics in the source code of Eclipse which is a Java based application contributing efficiently on the open source platform. A statistical analysis was pre applied on the set of code smells and metrics for finding the connection between the both. Later on, Multi Layer Perceptron model development on four versions of Eclipse has been made. Subsequently Area Under Curve (ROC) has been generated for class & method level code smells. The value of ROC in predicting code smells pointed towards the fact that Neural Network Multi Layer Perceptron model perform fair to good in determining the presence of code smells based on software metrics in Eclipse. Therefore from the results obtained it is concluded that smelly classes are predicted efficiently by software metric based code smells prediction model. The present study will be beneficial to the software development community to locate the refactoring areas and providing resources for testing. The results of empirical study also guide the development community by providing information relative to code smells and its types. The aim of this study is to provide statistical proof of linkage between metrics and code smells. The software metric based prediction � 2023 Scrivener Publishing LLC.
Insight into the liver dysfunction in COVID-19 patients: Molecular mechanisms and possible therapeutic strategies
(Baishideng Publishing Group Inc, 2023-04-12T00:00:00) Khullar, Naina; Bhatti, Jasvinder Singh; Singh, Satwinder; Thukral, Bhawana; Hemachandra Reddy, P.; Bhatti, Gurjit Kaur
As of June 2022, more than 530 million people worldwide have become ill with coronavirus disease 2019 (COVID-19). Although COVID-19 is most commonly associated with respiratory distress (severe acute respiratory syndrome), meta-analysis have indicated that liver dysfunction also occurs in patients with severe symptoms. Current studies revealed distinctive patterning in the receptors on the hepatic cells that helps in viral invasion through the expression of angiotensin-converting enzyme receptors. It has also been reported that in some patients with COVID-19, therapeutic strategies, including repurposed drugs (mitifovir, lopinavir/ritonavir, tocilizumab, etc.) triggered liver injury and cholestatic toxicity. Several proven indicators support cytokine storm-induced hepatic damage. Because there are 1.5 billion patients with chronic liver disease worldwide, it becomes imperative to critically evaluate the molecular mechanisms concerning hepatotropism of COVID-19 and identify new potential therapeutics. This review also designated a comprehensive outlook of comorbidities and the impact of lifestyle and genetics in managing patients with COVID-19. � The Author(s) 2023. Published by Baishideng Publishing Group Inc. All rights reserved.
Molecular dynamics and 3D-QSAR studies on indazole derivatives as HIF-1? inhibitors
(Taylor and Francis Ltd., 2022-03-23T00:00:00) Singh, Yogesh; Sanjay, Kulkarni Swanand; Kumar, Pradeep; Singh, Satwinder; Thareja, Suresh
Hypoxia-inducible factor (HIF) is a transcriptional factor which plays a crucial role in tumour metastasis thereby responsible for development of various forms of cancers. Indazole derivatives have been reported in the literature as potent HIF-1? inhibitor via interaction with key residues of the HIF-1? active site. Taking into consideration the role HIF-1? in cancer and potency of indazole derivative against HIF-1?; it was considered of interest to correlate structural features of known indazole derivatives with specified HIF-1? inhibitory activity to map pharmacophoric features through Three-dimensional quantitative structural activity relationship (3D-QSAR) and pharmacophore mapping. Field and Gaussian based 3D-QSAR studies were performed to realize the variables influencing the inhibitory potency of HIF-1? inhibitors. Field and Gaussian- based 3D-QSAR models were validated through various statistical measures generated by partial least square (PLS). The steric and electrostatic maps generated for both 3D-QSAR provide a structural framework for designing new inhibitors. Further; 3D-maps were also helpful in understanding variability in the activity of the compounds. Pharmacophore mapping also generates a common five-point pharmacophore hypothesis (A1D2R3R4R5_4) which can be employed in combination with 3D-contour maps to design potent HIF-1? inhibitors. Molecular docking and molecular dynamics (MD) simulation of the most potent compound 39 showed good binding efficiency and was found to be quite stable in the active site of the HIF-1? protein. The developed 3D-QSAR models; pharmacophore modelling; molecular docking studies along with the MD simulation analysis may be employed to design lead molecule as selective HIF-1? inhibitors for the treatment of Cancer. � 2022 Informa UK Limited, trading as Taylor & Francis Group.
Neural network based refactoring area identification in Software System with object oriented metrics
(Indian Society for Education and Environment, 2016) Kaur, Jaspreet; Singh, Satwinder
Objectives of the Study (a) To study previously designed models for identification of refactoring area in Object Oriented Software Systems. (b) To design a general framework or model that helps to easily identify the software code smells for a good quality of coding. (c) To identify the bad smells in the code with a design of neural network based model with the help of object-oriented metrics and further to predict the performance of the proposed model using various evaluation parameters of confusion matrix. Analysis/Methods: In this study, two different versions of Rhino (1.7r1 and 1.7r2) were taken as dataset. Object-Oriented metrics were taken as input data and the probability factor (occurrence or non-occurrence of a bad smell as output. Presence of a bad smell was considered as 1 and 0 means absence of bad smell. If there was at least one bad smell present in the code in a class, it was marked as smelly class. The tool used to extract the databases for collected object-oriented metrics and bad smells of these Rhino versions is PTIDEJ. Further, the data was tested on neural networks for different epochs to predict their performance. Findings: a) Bad Smell Analysis: Twelve design smells were considered to detect the presence of bad smell in code. If there was at least one bad smell present in the code in a class, it was marked as smelly class. b) Neural Network Model Table: Weight and bias factor for various predictors were calculated for different epochs (500, 1000, and 2000). It shows the weights assigned from input layer to hidden layer and from hidden layer to output neurons layer. After the training, the weights were tested on various datasets. C) Performance Tables and Graphs: In this, the Neural network proposed model was trained using different number of epochs to examine if the number of epochs used in training has any impact on the results or not. Further, the results for the accuracy of these models were shown. Novelty/Improvement: When the data was highly trained then the results were better. When the data was trained with 500 epochs, it was suitable for only with-in company projects but when the data was more trained than the model was also appropriate for cross projects. It was seen that when the data was trained with 1000 and 2000 epochs, the results of the proposed model were improved.
Object Oriented Metrics Based Empirical Model for Predicting �Code Smells� in Open Source Software
(Springer, 2023-01-03T00:00:00) Kaur, Sharanpreet; Singh, Satwinder
Refactoring is a technique which involves the change in internal layout of software system while keeping external features same as before. It aims at identification of smelly areas in code in order to apply patch. The presence of these smelly areas in code is also called �Code Smells� generally known as �Bad Smells.� Under the present study, the statistical relationship between software metrics, code smells and faults is investigated in open source systems. Eight versions of two open source projects � Eclipse and Tomcat are investigated in order to identify the effect on classes subject to code smells and faults than other classes. It was observed that in almost all releases of Tomcat, a good perceptiveness was revealed by area under ROC curve. While an excellent perceptiveness was collected by class level code smells for Eclipse, whereas a fair to good discrimination is generated by method level code smells by area under ROC curve. The results obtained from the study showed that software metrics model for smelly classes is helpful in predicting classes. � 2023, The Institution of Engineers (India).
Pharmacophore derived 3D-QSAR, molecular docking, and simulation studies of quinoxaline derivatives as ALR2 inhibitors
(Taylor and Francis Ltd., 2023-09-12T00:00:00) Singh, Yogesh; Kumar, Niraj; Kulkarni, Swanand; Singh, Satwinder; Thareja, Suresh
Aldose Reductase 2 (ALR2), a key enzyme of the polyol pathway, plays a crucial role in the pathogenesis of diabetic complications. Quinoxaline scaffold-based compounds have been identified as potential ALR2 inhibitors for the management of diabetic complications. In the present work, molecular dynamic simulation studies in conjugation with pharmacophore mapping and atom-based 3D-QSAR were performed on a dataset of 99 molecules in comparison with Epalrestat (reference) to mark the desirable structural features of quinoxaline analogs to generate a probable template for designing novel and effective ALR2 inhibitors. The most potent compound 81 was subjected to MD simulation studies and found to be stable, with better interactions with the binding pocket as compared to Epalrestat. The MM-GBSA and MM-PBSA calculations showed that compound 81 possessed binding free energies of ?35.96 and ?4.92 kcal/mol, respectively. Atom-based 3D-QSAR yielded various pharmacophoric features with excellent statistical measures, such as correlation coefficient (R 2 value), F-value (Fischer ratio), Q 2 value (cross-validated correlation coefficient), and Pearson�s R-value for training and test sets. Furthermore, the pharmacophore mapping provided a five-point hypothesis (AADRR) and docking analysis revealed the active ligand-binding orientations on the active site�s amino acid residues TYR 48, HIE 110, TRP 111, and TRP 219. The results of this study will help in designing potent inhibitors of ALR2 for the management of diabetic complications. Communicated by Ramaswamy H. Sarma. � 2023 Informa UK Limited, trading as Taylor & Francis Group.
Proposed Model for Context Topic Identification of English and Hindi News Article Through LDA Approach with NLP Technique
(Springer, 2021-08-14T00:00:00) Srivastav, Anukriti; Singh, Satwinder
According to the survey, India has the world's second-largest newspaper market, with more than 100�K newspaper outlets, approx 240 million circulation, and 1300 million subscribers or readers. The topic modeling work is increasing day by day, and researchers have published multiple topic modeling papers and have implemented them in different areas like software engineering, political science and medical, etc. LDA topic modeling is used in this research because it has been introduced successfully for topic modeling and classification and it measures the probability of a text-dependent on the bag-of-words scheme without considering the word series. LDA is a common topic modeling algorithm with excellent implementation in the Gensim Python package. However, the challenge is how to extract good quality topics that are simple, separated, and meaningful. The purpose of this research deals with finding the main topics of the same category news articles which are in two different languages (Hindi and English) and then classifying these different language news topics with similarity measurement. In this research, the corpus is constructed with bigram. To achieve the research goal, we have to first build a headline and link extractor that scrap the top news from Google News feeds for both English and Hindi languages (Google News collects news stories that have appeared on different news website which is already accessible in 35 languages over the last 30�days) and then analyses which two news headlines are similar. � 2021, The Institution of Engineers (India).
A study on fake news subject matter, presentation elements, tools of detection, and social media platforms in India
(Center for Asian Public Opinion Research and Collaboration Initiative, 2021-02-28T00:00:00) Kanozia, Rubal; Arya, Ritu; Singh, Satwinder; Narula, Sumit; Ganghariya, Garima
This research article attempts to understand the current situation of fake news on social media in India. The study focused on four characteristics of fake news based on four research questions: subject matter, presentation elements of fake news, debunking tool(s) or technique(s) used, and the social media site on which the fake news story was shared. A systematic sampling method was used to select a sample of 90 debunked fake news stories from two Indian fact-checking websites, Alt News and Factly, from December 2019 to February 2020. A content analysis of the four characteristics of fake news stories was carefully analyzed, classified, coded, and presented. The results show that most of the fake news stories were related to politics in India. The majority of the fake news was shared via a video with text in which narrative was changed to mislead users. For the largest number of debunked fake news stories, information from official or primary sources, such as reports, data, statements, announcements, or updates were used to debunk false claims. � 2021, Center for Asian Public Opinion Research and Collaboration Initiative. All rights reserved.
A survey on near-human conversational agents
(King Saud bin Abdulaziz University, 2021-11-10T00:00:00) Singh, Satwinder; Beniwal, Himanshu
Conversational AI intends for machine-human interactions to appear and feel more natural and inclined to communicate in a near-human context. Chatbots, also known as conversational agents, are typically divided into two use-cases: task-oriented bots and social friend-bots. Task-oriented bots are often used to do activities such as answering questions or solving basic queries. Furthermore, social-friend-bots are designed to communicate like humans, where the user can speak freely and the bot answers organically while maintaining the conversation's ambience. This paper analyses recent works in the conversational AI domain examining the exclusive methodologies, existing frameworks or tools, evaluation metrics, and available datasets for building robust conversational agents. Finally, a mind-map encompassing all the stated elements and qualities of chatbots is created. � 2021 The Authors