In this paper, we measured the privacy budgets and utilities of different differentially private mechanisms combined with different machine learning models that forecast traffic congestion at future timestamps. We expected the ANNs combined with the Staircase mechanism to perform the best with every value in the privacy budget range, especially with the medium high values of the privacy budget. In this study, we used the Autoregressive Integrated Moving Average (ARIMA) and neural network models to forecast and then added differentially private Laplacian, Gaussian, and Staircase noise to our datasets. We tested two real traffic congestion datasets, experimented with the different models, and examined their utility for different privacy budgets. We found that a favorable combination for this application was neural networks with the Staircase mechanism. Our findings identify the optimal models when dealing with tricky time series forecasting and can be used in non-traffic applications like disease tracking and population growth.

Here, seeking to understand the effects of social media in relation to social media fatigue and/or overload in recent years, the authors used various linear models to assess the results of a survey of 27 respondents. Their results showed that increased duration of use of social media did not necessarily lead to fatigue, suggesting that quality may be more important than quantity. They also considered the purpose of an individual's social media usage as well as their engagement behavior during the COVID-19 pandemic.

Osteosarcoma is a type of bone cancer that affects young adults and children. Early diagnosis of osteosarcoma is crucial to successful treatment. The current methods of diagnosis, which include imaging tests and biopsy, are time consuming and prone to human error. Hence, we used deep learning to extract patterns and detect osteosarcoma from histological images. We hypothesized that the combination of two different technologies (transfer learning and data augmentation) would improve the efficacy of osteosarcoma detection in histological images. The dataset used for the study consisted of histological images for osteosarcoma and was quite imbalanced as it contained very few images with tumors. Since transfer learning uses existing knowledge for the purpose of classification and detection, we hypothesized it would be proficient on such an imbalanced dataset. To further improve our learning, we used data augmentation to include variations in the dataset. We further evaluated the efficacy of different convolutional neural network models on this task. We obtained an accuracy of 91.18% using the transfer learning model MobileNetV2 as the base model with various geometric transformations, outperforming the state-of-the-art convolutional neural network based approach.

This article discusses Alopecia areata, an autoimmune disorder causing sudden hair loss due to the immune system mistakenly attacking hair follicles. The article introduces the use of deep learning (DL) techniques, particularly convolutional neural networks (CNN), for classifying images of healthy and alopecia-affected hair. The study presents a comparative analysis of newly optimized CNN models with existing ones, trained on datasets containing images of healthy and alopecia-affected hair. The Inception-Resnet-v2 model emerged as the most effective for classifying Alopecia Areata.

Neural machine translation (NMT) is a software that uses neural network techniques to translate text from one language to another. However, one of the most famous NMT models—Google Translate—failed to give an accurate English translation of a famous Korean nursery rhyme, "Airplane" (비행기). The authors fine-tuned a pre-trained model first with a dataset from the lyrics domain, and then with a smaller dataset containing the rhythmical properties, to teach the model to translate rhythmically accurate lyrics. This stacked fine-tuning method resulted in an NMT model that could maintain the rhythmical characteristics of lyrics during translation while single fine-tuned models failed to do so.

Here, the authors investigated the most efficient way to position magnets to hold the most pieces of paper on the surface of a refrigerator. They used a regression model along with an artificial neural network to identify the most efficient positions of four magnets to be at the vertices of a rectangle.

Ant colony optimization algorithms simulate ants moving from point to point on a graph and coordinate their actions, similar to ants laying down pheromones to strengthen a path as it is used more frequently. These ACO algorithms can be applied to the classic traveling salesman problem, which aims to determine the lowest-cost path through a given set of points on a graph. In this study, a novel multiple-colony system was developed that uses multiple simulated ant colonies to generate improved solutions to the traveling salesman problem.

Based on the success of deep learning, recent works have attempted to develop a waste classification model using deep neural networks. This work presents federated learning (FL) for a solution, as it allows participants to aid in training the model using their own data. Results showed that with less clients, having a higher participation ratio resulted in less accuracy degradation by the data heterogeneity.

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder and is difficult to diagnose in young children. Here magnetoencephalography was used to compare the brain activity in patients with ASD to patients in a control group. The results show that patients with ASD have a high level of activity in different areas of the brain than those in the control group.

In this study, the authors seek to improve a machine learning algorithm used for image classification: identifying male and female images. In addition to fine-tuning the classification model, they investigate how accuracy is affected by their changes (an important task when developing and updating algorithms). To determine accuracy, a set of images is used to train the model and then a separate set of images is used for validation. They found that the validation accuracy was close to the training accuracy. This study contributes to the expanding areas of machine learning and its applications to image identification.