Journal of Big Data and Artificial Intelligence

A New Era of Artificial Intelligence Begins – Where Will it Lead Us?

Jim Samuel, Abhishek Tripathi, Ensela Mema — Sun, 07 Jan 2024 00:00:00 -0500

In this Editorial, we highlight the emerging dominance of AI + Big Data, and here are some excerpts : We have entered into the age of Artificial Intelligence (AI). Everything around us is becoming artificially intelligent: from business applications to healthcare, education to finance and governance to art, music and entertainment. The fact that AI has gripped public attention is evident from the steep rise in public engagement with artificial intelligence applications, explosive increase in news media coverage of AI, increasing volumes of social media posts and the mushrooming of a range of AI ecosystem initiatives. We at JBDAI (formerly JBDTP) hope to encourage and foster much high quality research, rigor and innovative thought leadership on big data and artificial intelligence in the years ahead, supporting human well-being, the sustainability of our natural resources and balanced societal progress – please contribute to JBDAI and be a part of this exciting intellectual adventure!

In Memory of Dr. David Belanger

Tue, 09 Jan 2024 00:00:00 -0500

Sadly, our dear colleague, Dr. David Belanger, passed away in November last year. David was a founding member of the New Jersey Big Data Alliance (NJBDA)—an alliance of New Jersey academic institutions and corporations that aims to promote Big Data education and research in New Jersey, the parent organization of this journal. “Through the last decade, as our organization grew and expanded its programs, he provided brilliant insight and guidance on our direction, offering suggestions in his thoughtful way and always ready to collaborate. David will be greatly missed,” said Margaret Brennan-Tonetta, NJBDA’s past president and co-founder. At NJBDA, he was most recently Vice President of the Entrepreneurship Committee. David was an internationally known authority on Big Data and data governance. We at NJBDA and JBDAI will continue to remember David as a gentle scholar who cared for people. A colleague ﬁttingly remembered David as being “the kindest scientist of our time.”

BERT based Blended approach for Fake News Detection

Satish Mahadevan sr, Shafqaat Ahmad — Sun, 07 Jan 2024 00:00:00 -0500

This paper presents a new approach for detecting fake news on social media. Previous works in this domain have demonstrated that context is an important factor when attempting to distinguish subtle differences within text. Fake news itself presents different level of difficulty due the vast similarity that exists between genuine and fake news contents. Therefore, we propose a collaborative approach which uses probabilistic fusion strategy to combine the knowledge gained from modelling two language models, BERT-LSTM and BERT-CNN. To achieve the fusion, we exploit the Bayesian method. Our experiments are conducted on two fake news detection datasets. The detection accuracy attained in these experiments attest to the efficiency of the proposed method, as our approach is very competitive compared to the state-of-the-art methods.

Investment under Uncertainty: The Role of Inventory Dynamics

CHUANQIAN ZHANG, Xue Cui, Sudipto Sarkar — Sun, 07 Jan 2024 00:00:00 -0500

Finished-good inventory is very common under market uncertainty. We build a continuous-time model to study how the inventory will impact firm value and investment decisions. Our model shows that the value of a company following the optimal inventory policy can be significantly higher than the traditional non-inventory company, particularly if the inventory-holding cost is not large. This premium becomes small as holding cost is increased, and large when demand is volatile, and when price elasticity is large. We also show that the optimal investment size can be significantly larger than the traditional no-inventory firm, particularly when the inventory-holding cost is low, demand volatility is high, and price elasticity is low. This paper develops a simulation algorithm to solve iterative optimization problem in a path-dependent economy.

Crime Frequency During COVID - 19 and Black Lives Matter Protests

Aylin Kosar, Mehmet Turkoz — Sun, 07 Jan 2024 00:00:00 -0500

The COVID-19 disrupted the daily life of individuals within the United States and around the world when government restrictions were put into place. During the pandemic restrictions, social unrest took place after the death of George Floyd. Our objective is to study the crime rate during the pandemic and social unrest that took place after the death of George Floyd. We used data from four cities that were heavily affected with the pandemic and social unrest: Seattle, San Francisco, Los Angeles, and Philadelphia. Holt-Winters and SARIMA models were used to see if there was any change of crime during the pandemic and social unrest in addition to before and after the social unrest. Los Angeles had the lowest crime frequency out of the four cities while Philadelphia had the highest. All Holt-Winters models and SARIMA models showed around January 2020, during when the first case of COVID-19 occurred, crime was the same for all four cities except for Philadelphia where crime had dropped for a particular time until it increased again. There was no clear evidence to suggest that crime was affected during the COVID-19 pandemic and the social unrest during the protests.

Machine Learning Study: Identification of Skin Diseases for Various Skin Types Using Image Classification.

Sun, 07 Jan 2024 00:00:00 -0500

Increased machine learning methods have helped improvise human interaction with digital devices which helps in skin disease identification, prediction, and classification by employing algorithms. Image classification for skin disease application algorithms can detect caucasian skin tones but poorly performs when analyzing other skin colors. In this research, a deep learning algorithm was used to address the problem that other applications perform poorly with the classification of skin disease types.

Convolutional Neural Network (CNN), a machine-learning algorithm was used to classify images and add the predicted images within the data set. The images in the data set covered a lot of patient factors such as age, sex, disease site (hand, feet, head, nails, etc.), skin color (white, yellow, brown, black) and different periods of lesions (early, middle, or late). Multiple private applications can detect skin diseases during the analysis. For the darker color skin population, the performance was poor, and skin cancer detection was not possible even with the help of image recognition.

This research aims to conduct an analysis of visual searches within skin-related health searches to identify opportunities to provide digital health consumers with visual search results that are more representative of America’s diverse populations.

Are Emotions Conveyed Across Machine Translations? Establishing an Analytical Process for the Effectiveness of Multilingual Sentiment Analysis with Italian Text

Richard Anderson, Carmela Scala, Jim Samuel, Vivek Kumar, Parth Jain — Sun, 07 Jan 2024 00:00:00 -0500

Abstract Natural language processing (NLP) is being widely used globally for a variety of value-creation tasks ranging from chat-bots and machine translations to sentiment and topic analysis and multilingual large language models (LLMs). However, most of the advances are initially implemented within the English language framework, and it takes time and resources to develop comparable resources in other languages. The advances in machine translations have enabled the rapid and effective conversion of content in global languages into English and vice-versa. This creates potential opportunities to apply English language NLP methods and tools to other languages via machine translations. However, although this idea is powerful, it needs to be validated and processes and best practices need to be developed and kept updated. The present research is an effort to contribute to the development of best practices and an evaluation framework. We present a systematic and repeatable state-of-the-art process to evaluate the viability of applying English language sentiment analysis tools to Italian text by using multiple English language machine translation mechanisms such that it can be easily extended to other languages.