International journal of information technology : an official journal of Bharati Vidyapeeth's Institute of Computer Applications and Management最新文献

Biometric features are useful for unique identification, authentication, and security applications. Among all biometric features, fingerprints are the most commonly used because they contain ridges and valleys. There are challenges in recognizing child or infant fingerprints since the ridges are not mature as the hands are covered with a white substance and acquisition of fingerprint images is difficult. In the context of COVID-19 pandemic, contactless fingerprint acquisition gains importance as it is not infectious especially with children. In this study, a Convolutional Neural Network (CNN) based children recognition system named Child-CLEF, that uses Contact-Less Children Fingerprint (CLCF) dataset acquired using a mobile phone-based scanner is proposed. The quality of captured fingerprint images is enhanced using a hybrid image enhancement method. Furthermore, the minutiae features are extracted using the proposed Child-CLEF Net model and the identification of children is made using a matching algorithm. The proposed system is tested with a self-captured children fingerprint dataset, CLCF and publicly available PolyU fingerprint dataset. It is found that the proposed system outperforms the existing fingerprint recognition systems in terms of accuracy and equal error rate.

The present communication introduces a new discriminant measure coined as R-norm q-rung picture fuzzy discriminant information measure which is more generalized in nature and has the capability to handle more flexibility inherited in the inexact information. The notion of q-rung picture fuzzy set (q-RPFS) has an integrated advantage of picture fuzzy set and q-rung orthopair fuzzy set with flexibility of qth level relations. The proposed parametric measure is then applied in the conventional "technique for order preference by similarity to the ideal solution (TOPSIS) method" for solving a green supplier selection problem. The numerical illustration to exhibit the proposed methodology for the green supplier selection problem has been presented in an empirical form to establish the consistency of the model. Also, the advantageous features of the proposed scheme in the setup of impreciseness have been discussed.

The construction sector in a rapidly developing country like India is a very unorganized sector. A large number of workers were affected and hospitalized during the pandemic. This situation is costing the sector heavily in several respects. This research study was conducted as part of using machine learning algorithms to improve construction company health and safety policies. LOS (length of stay) is used to predict how long a patient will stay in a hospital. Predicting LOS is very useful not only for hospitals, but also for construction companies to measure resources and reduce costs. Predicting LOS has become an important step in most hospitals before admitting patients. In this post, we used the Medical Information Mart for Intensive Care(MIMIC III) dataset and applied four different machine learning algorithms: decision tree classifier, random forest, Artificial Neural Network (ANN), and logistic regression. First, I performed data pre-processing to clean up the dataset. In the next step, we performed function selection using the Select Best algorithm with an evaluation function of chi2 to perform hot coding. We then performed a split between training and testing and applied a machine learning algorithm. The metric used for comparison was accuracy. After implementing the algorithms, the accuracy was compared. Random forest was found to perform best at 89%. Afterwards, we performed hyperparameter tuning using a grid search algorithm on a random forest to obtain higher accuracy. The final accuracy is 90%. This kind of research can help improve health security policies by introducing modern computational techniques, and can also help optimize resources.

The rise of cryptocurrency, especially bitcoin, has opened up a lot of doors in the world of Financial Technology (FinTech) by attracting investors, media, and financial industry regulators. Bitcoin operates on blockchain technology, and its value is not a determinant of the value of a tangible asset, an organisation, or a country's economy. Instead, it relies on an encryption technique that allows tracking of all transactions. Globally, over $2 trillion has been generated through cryptocurrency trading. Due to these financial prospects, the youths in Nigeria have cashed in on this virtual currency to create employment and wealth. This research investigates the adoption and sustainability of bitcoin and blockchain in Nigeria. A survey method with a non-probability purposive sampling technique and a homogeneous approach was employed to collect 320 responses via an online survey. Descriptive and correlational analysis in IBM SPSS version 25 was used to analyse the collected data. According to the findings, bitcoin is the most popular cryptocurrency, with 97.5% acceptance, and is expected to be the leading virtual currency in the next five years. The research findings will help researchers and authorities comprehend the need for cryptocurrency adoption, leading to its sustainability.

Fake news on social media has become a growing concern due to its potential impact on shaping public opinion. The proposed Debunking Multi-Lingual Social Media Posts using Deep Learning (DSMPD) approach offers a promising solution to detect fake news. The DSMPD approach involves creating a dataset of English and Hindi social media posts using web scraping and Natural Language Processing (NLP) techniques. This dataset is then used to train, test, and validate a deep learning-based model that extracts various features, including Embedding from Language Models (ELMo), word and n-gram counts, Term Frequency-Inverse Document Frequency (TF-IDF), sentiments, polarity, and Named Entity Recognition (NER). Based on these features, the model classifies news items into five categories: real, could be real, could be fabricated, fabricated, or dangerously fabricated. To evaluate the performance of the classifiers, the researchers used two datasets comprising over 45,000 articles. Machine learning (ML) algorithms and Deep learning (DL) model are compared to choose the best option for classification and prediction.

In this research study, researcher tries to understand how Over-the-top platforms like Netflix categorically utilize aesthetic visual analysis (AVA); an image selection tool to reduce time and increase efficacy, through a parametric analysis of AVA to optimise Netflix performance. This research paper tries to answer all the questions related to the how the database of aesthetic visual analysis (AVA), an image selection tool works better or more like humans. To further substantiate the popularity of Netflix, a real time data of 307 respondents who use OTT platforms in Delhi was collected to determine whether Netflix in fact is or not the market leader. 63.8% of them selected Netflix as their top option.

Overcrowding in hospitals in Vietnam has caused many disadvantages in receiving and treating patients. Especially at the stage of receiving and diagnosing procedures taking patients to the treatment departments in the hospital takes up much time. This study proposes a text-based disease diagnosis using text processing techniques (such as Bag of Words, Term Frequency- Inverse Document Frequency, and Tokenizer) combined with classifiers (such as Random Forests (RF), Multi-Layer Perceptron (MLP), Embeddings and Bidirectional Long Short-term memory (LSTM)) on symptoms. As observed from the results, deep Bidirectional LSTM can reach 0.982 in AUC in the classification of 10 diseases on 230,457 samples of pre-diagnosis collected from Vietnam hospitals used in the training and testing phases. The proposed approach is expected to provide a way to automate patient flow in hospitals to improve healthcare in the future.

The corona virus (COVID-19) pandemic has impacted industries across the globe. Lockdown was imposed to curb the spread of the deadly virus. This resulted in closure of the factories and manufacturing units. Few sectors switched to work from home (WFH) for the first time. The present study aims to understand and analyze the way in which Information Technology (IT) sector communicated on Twitter during the pandemic. The top ten IT companies in India were selected on the basis of net sales. Qualitative data analysis was employed to extract tweets, understand and analyze them. Tweets were extracted from the official Twitter handles of these top ten IT companies using N-Capture extension tool of NVivo 12 software from April 1, 2020 to April 30, 2021. To get insights out of collected data, Word Cloud, TreeMap and Sentiment Analysis of tweets were carried out using NVivo 12 software. The research found that IT companies focussed on digital transformation, business development, customer satisfaction and enriching customer experience, new product development for healthcare and insurance and organizational resilience. They also focussed on effective communication through Twitter in times of crisis. Most of the companies tweeted moderately positive. Very small numbers of tweets were found to be very negative.

The Deep Convolutional Neural Network (DCNN) classification models are being tremendously used across many research fields including medical science for image classification. The accuracy of the model and reliability on the results of the model are the key attributes which determine whether a particular model should be used for a specific application or not. A highly accurate model is always desirable for all applications of machine learning as well as deep learning. This paper presents a DCNN based heterogeneous ensemble approach where all DCNN models can be trained on a single dataset and each model can contribute of towards the final output of the ensemble model. The contribution of each model is weighted according to its individual accuracy on the given dataset. Models with higher accuracy has higher contribution in the final output of ensemble model, whereas the models with lower accuracy has lower contribution. This approach, when tested on two different X-ray images datasets of Covid-19, has confirmed the significant increase in 3-class accuracy as compared to the models in literature.

Healthcare is on top of the agenda of all governments in the world as it is related to the well-being of the people. Naturally, this domain has attracted the attention of many researchers globally, who have studied the development of its different phases, including E-Health and the Internet of Health Things (IoHT). In this paper, the difference between the recent concepts of healthcare (E-health, M-Health, S-Health, I-Health, U-Health, and IoHT/IoMT) is analyzed based on the main services, applications, and technologies in each concept. The paper has also studied the latest developments in IoHT, which are linked to existing phases of development. A classification of groups of services and constituents of IoHT, linked to the latest technologies, is also provided. In addition, challenges, and future scope of research in this domain concerning the wellbeing of the people in the face of ongoing COVID-19 and future pandemics are explored.