Automatic recognition of human activities using wearable sensors remains a challenging problem due to high variability in inter-person gait and movements. Moreover, finding the best on-body location for a wearable sensor is also critical though it provides valuable context information that can be used for accurate recognition. This paper addresses the problem of classifying motion signals generated by multiple wearable sensors for the recognition of human activity and localisation of the wearable sensors. Unlike existing methods that used the raw accelerometer and gyroscope signals for extracting time and frequency-based features for activity inference, we propose to create frequency images for the raw signals and show this representation to be more robust. The frequency image sequences are generated from the accelerometer and gyroscope signals from seven different body parts. These frequency images serve as the input to our proposed two-stream Convolutional Neural Networks (CNN) for predicting the human activity and the location of the sensor generating the activity signal. We show that the complementary information collected by both accelerometer and gyroscope sensors can be leveraged to develop an effective classifier that can accurately predict the performed human activity. We evaluate the performance of the proposed method using the cross-subjects approach and show that it achieves an impressive F1-score of 0.90 on a publicly available real-world human activity dataset. This performance is superior to that reported by another state-of-the-art method on the same dataset. Moreover, we also experimented with the datasets from different body locations to predict the best position for the underlying task. We show that shin and waist are the best places on the body for placing sensors and this could help other researchers to collect higher quality activity data. We plan to publicly release the generated frequency images from all sensor positions and activities and our implementation code with the publication.

Waste management is an essential societal issue, and the classical and manual waste auditing methods are hazardous and time-consuming. In this paper, we introduce a novel method for waste detection and classification to address the challenges of waste management. The method uses a collection of deep neural networks to allow for accurate waste detection, classification, and waste size quantification. The trained neural network model is integrated into a mobile-based application for trash geotagging based on images captured by users on their smartphones. The tagged images are then connected to the cleaners’ database, and the nearest cleaners are notified of the waste. The experimental results using publicly available datasets show the effectiveness of the proposed method in terms of detection and classification accuracy. The proposed method achieved an accuracy of at least 90%, which surpasses that reported by other state-of-the-art methods on the same datasets.

Students’ feedback is pertinent in measuring the quality of the educational process. For example, by applying lexicon-based sentiment analysis to students’ open-ended course feedback, we can detect not only their sentiment orientation (positive, negative, or neutral) but also their emotional valences, such as anger, anticipation, disgust, fear, joy, sadness, surprise, or trust. However, most currently used assessment tools cannot effectively measure emotional engagement, such as interest level, enjoyment, support, curiosity, and sense of belonging. Moreover, none of those tools utilize Bloom’s taxonomy for students’ learning-level assessment. In this work, we develop a user-friendly application based on NLP to help the teachers understand the students’ perception of their learning by analyzing their open-ended feedback. This allows us to examine the sentiment and the embedded emotions using a customized dictionary of emotions related to education. The application can also classify the students’ emotions according to Bloom’s taxonomy. We believe our application will help teachers improve their course delivery.

Diabetic Retinopathy is a serious complication arising in diabetes afflicted patients. Its effective treatment depends on early detection, and the course of action varies decisively with the intensity of the affliction. Computer-aided diagnosis helps to detect not only the presence or absence of the disease but also the severity, making it easier for ophthalmologists to construct a treatment plan. Diabetic retinopathy grading is the task of classifying images of the eye's fundus of diabetic patients into 5 different grades ranging from 0-4 based on the severity of the disease. In this work, we propose a deep neural network architecture to address the grading problem. The method utilizes an additional attention layer in the neural network model to capture the spatial relationship between the region of interest in the images during the training process to better discriminate between the different severity stages of the disease. Also, we analyze the impact of different image processing techniques on the classification results. We assessed the performance of our proposed method using a dataset of eye fundus images and obtained a classification accuracy of 89.20% on average. This performance surpasses that reported for other state-of-the-art methods on the same dataset. The effectiveness of the proposed method will facilitate the procedural workflow of identifying severe cases of diabetic retinopathy

In the modern era of Information Technology, Cloud Computing (CC) technology plays a prominent role in academics and industry, and even health care organiz In the modern era of Information Technology, Cloud Computing (CC) technology plays a prominent role in academics and industry, and even health care organiz