Deep Learning-Based Water Crystal Classification

Much of the earth’s surface is covered by water. As was pointed out in the 2020 edition of the World Water Development Report, climate change challenges the sustainability of global water resources, so it is important to monitor the quality of water to preserve sustainable water resources. Quality of water can be related to the structure of water crystal, the solid-state of water, so methods to understand water crystals can help to improve water quality. As a first step, a water crystal exploratory analysis has been initiated with the cooperation with the Emoto Peace Project (EPP). The 5K EPP dataset has been created as the first world-wide small dataset of water crystals. Our research focused on reducing the inherent limitations when fitting machine learning models to the 5K EPP dataset. One major result is the classification of water crystals and how to split our small dataset into several related groups. Using the 5K EPP dataset of human observations and past research on snow crystal classification, we created a simple set of visual labels to identify water crystal shapes, in 13 categories. A deep learning-based method has been used to automatically do the classification task with a subset of the label dataset. The classification achieved high accuracy when using a fine-tuning technique.

A Novel Multistage Transfer Learning for Ultrasound Breast Cancer Image Classification

Breast cancer diagnosis is one of the many areas that has taken advantage of artificial intelligence to achieve better performance, despite the fact that the availability of a large medical image dataset remains a challenge. Transfer learning (TL) is a phenomenon that enables deep learning algorithms to overcome the issue of shortage of training data in constructing an efficient model by transferring knowledge from a given source task to a target task. However, in most cases, ImageNet (natural images) pre-trained models that do not include medical images, are utilized for transfer learning to medical images. Considering the utilization of microscopic cancer cell line images that can be acquired in large amount, we argue that learning from both natural and medical datasets improves performance in ultrasound breast cancer image classification. The proposed multistage transfer learning (MSTL) algorithm was implemented using three pre-trained models: EfficientNetB2, InceptionV3, and ResNet50 with three optimizers: Adam, Adagrad, and stochastic gradient de-scent (SGD). Dataset sizes of 20,400 cancer cell images, 200 ultrasound images from Mendeley and 400 ultrasound images from the MT-Small-Dataset were used. ResNet50-Adagrad-based MSTL achieved a test accuracy of 99 ± 0.612% on the Mendeley dataset and 98.7 ± 1.1% on the MT-Small-Dataset, averaging over 5-fold cross validation. A p-value of 0.01191 was achieved when comparing MSTL against ImageNet based TL for the Mendeley dataset. The result is a significant improvement in the performance of artificial intelligence methods for ultrasound breast cancer classification compared to state-of-the-art methods and could remarkably improve the early diagnosis of breast cancer in young women.

An evaluation of approaches for enhancing inductive learning with a transductive view

Availability of enough labeled data is a challenge for most inductive learners who try to generalize based on limited labeled dataset. A traditional semi-supervised approach for the same problem attempts to approach it by methods such as wrapping multiple inductive learners on derived pseudo-labels, unsupervised feature extraction or suitable modification of the objective function. In this work, a simple approach is adopted whereby an inductive learner is enhanced by suitably enabling it with a transductive view of the data. The experiments, though conducted on a small dataset, successfully provide few insights i.e. transductive view benefits an inductive learner, a transductive view that considers both attribute and relations is more effective than one that considers either attributes or relations and graph convolution based embedding algorithms effectively captures the information from transductive views compared to popular knowledge embedding approaches.

Analyzing Graduation Project Ideas by using Machine Learning

The graduation projects (GP) are important because it reflects the academic profile and achievement of the students. For many years’ graduation projects are done by the information technology department students. Most of these projects have great value, and some were published in scientific journals and international conferences. However, these projects are stored in an archive room haphazardly and there is a very small part of it is a set of electronic PDF files stored on hard disk, which wastes time and effort and cannot benefit from it. However, there is no system to classify and store these projects in a good way that can benefit from them. In this paper, we reviewed some of the best machine learning algorithms to classify text “graduation projects”, support vector machine (SVM) algorithm, logistic regression (LR) algorithm, random forest (RF) algorithm, which can deal with an extremely small amount of dataset after comparing these algorithms based on accuracy. We choose the SVM algorithm to classify the projects. Besides, we will mention how to deal with a super small dataset and solve this problem.

Vision Transformer-Based Tailing Detection in Videos

Tailing is defined as an event where a suspicious person follows someone closely. We define the problem of tailing detection from videos as an anomaly detection problem, where the goal is to find abnormalities in the walking pattern of the pedestrians (victim and follower). We, therefore, propose a modified Time-Series Vision Transformer (TSViT), a method for anomaly detection in video, specifically for tailing detection with a small dataset. We introduce an effective way to train TSViT with a small dataset by regularizing the prediction model. To do so, we first encode the spatial information of the pedestrians into 2D patterns and then pass them as tokens to the TSViT. Through a series of experiments, we show that the tailing detection on a small dataset using TSViT outperforms popular CNN-based architectures, as the CNN architectures tend to overfit with a small dataset of time-series images. We also show that when using time-series images, the performance of CNN-based architecture gradually drops, as the network depth is increased, to increase its capacity. On the other hand, a decreasing number of heads in Vision Transformer architecture shows good performance on time-series images, and the performance is further increased as the input resolution of the images is increased. Experimental results demonstrate that the TSViT performs better than the handcrafted rule-based method and CNN-based method for tailing detection. TSViT can be used in many applications for video anomaly detection, even with a small dataset.

Detecting Excessive Daytime Sleepiness With CNN And Commercial Grade EEG

Epworth sleepiness scale is a self-assessment method in sleep medicine that has been proven to be a good predictor of obstructive sleep apnea. However, the over-reliance of the method making the process not socially distancing friendly enough in response to a global covid-19 pandemic. A study states that the Epworth sleepiness scale is correlated with the brainwave signal that commercial-grade EEG can capture. This study tried to train a classifier powered by CNN and deep learning that could perform as well as the Epworth with the objectiveness of brainwave signal. We test the classifier using the 20 university student using the Epworth sleepiness test beforehand. Then, we put the participant in 10 minutes EEG session, downsampling the data for normalization purposes and trying to predict the outcome of the ESS in respect of their brainwave state. The AI predict the reaching 65% of accuracy and 81% of sensitivity with just under 100.000 dataset which is excellent considering small dataset although this still have plenty room for improvement.

Fusion-Based Semantic Segmentation Using Deep Learning Architecture in Case of Very Small Training Dataset

Semantic segmentation is a pre-processing step in computer vision-based applications. It is the task of assigning a predefined class label to every pixel of an image. Several supervised and unsupervised algorithms are available to classify pixels of an image into predefined object classes. The algorithms, such as random forest and SVM are used to obtain the semantic segmentation. Recently, convolutional neural network (CNN)-based architectures have become popular for the tasks of object detection, object recognition, and segmentation. These deep architectures perform semantic segmentation with far better accuracy than the algorithms that were used earlier. CNN-based deep learning architectures require a large dataset for training. In real life, some of the applications may not have sufficient good quality samples for training of deep learning architectures e.g. medical applications. Such a requirement initiated a need to have a technique of effective training of deep learning architecture in case of a very small dataset. Class imbalance is another challenge in the process of training deep learning architecture. Due to class imbalance, the classifier overclassifies classes with large samples. In this paper, the challenge of training a deep learning architecture with a small dataset and class imbalance is addressed by novel fusion-based semantic segmentation technique which improves segmentation of minor and major classes.

ASER: Arabic Speech Emotion Recognition Employing Wav2vec2.0 and HuBERT Based on BAVED Dataset

Recently, there have been tremendous research outcomes in the fields of speech recognition and natural language processing. This is due to the well-developed multilayers deep learning paradigms such as wav2vec2.0, Wav2vecU, WavBERT, and HuBERT that provide better representation learning and high information capturing. Such paradigms run on hundreds of unlabeled data, then fine-tuned on a small dataset for specific tasks. This paper introduces a deep learning constructed emotional recognition model for Arabic speech dialogues. The developed model employs the state of the art audio representations include wav2vec2.0 and HuBERT. The experiment and performance outcomes of our model overcome the previous known results.