The federated learning technique (FL) supports the collaborative training of machine learning and deep learning models for edge network optimization. Although a complex edge network with heterogeneous devices having different constraints can affect its performance, this leads to a problem in this area. Therefore, some research can be seen to design new frameworks and approaches to improve federated learning processes. The purpose of this study is to provide an overview of the FL technique and its applicability in different domains. The key focus of the paper is to produce a systematic literature review of recent research studies that clearly describes the adoption of FL in edge networks. The search procedure was performed from April 2020 to May 2021 with a total initial number of papers being 7546 published in the duration of 2016 to 2020. The systematic literature synthesizes and compares the algorithms, models, and frameworks of federated learning. Additionally, we have presented the scope of FL applications in different industries and domains. It has been revealed after careful investigation of studies that 25% of the studies used FL in IoT and edge-based applications and 30% of studies implement the FL concept in the health industry, 10% for NLP, 10% for autonomous vehicles, 10% for mobile services, 10% for recommender systems, and 5% for FinTech. A taxonomy is also proposed on implementing FL for edge networks in different domains. Moreover, another novelty of this paper is that datasets used for the implementation of FL are discussed in detail to provide the researchers an overview of the distributed datasets, which can be used for employing FL techniques. Lastly, this study discusses the current challenges of implementing the FL technique. We have found that the areas of medical AI, IoT, edge systems, and the autonomous industry can adapt the FL in many of its sub-domains; however, the challenges these domains can encounter are statistical heterogeneity, system heterogeneity, data imbalance, resource allocation, and privacy.

Opinion Mining or Sentiment Analysis is the process of mining emotions, attitudes, and opinions automatically from speech, text, and database sources through Natural Language Processing (NLP). Opinions can be given on anything. It may be a product, feature of a product or any sentiment view on a product. In this research, Mobile phone products reviews, fetched from Amazon.com, are mined to predict customer rating of the product based on its user reviews. This is performed by the sentiment classification of unlocked mobile reviews for the sake of opinion mining. Different opinion mining algorithms are used to identify the sentiments hidden in the reviews and comments for a specific unlocked mobile. Moreover, a performance analysis of Sentiment Classification algorithms is performed on the data set of mobile phone reviews. Results yields from this research provide the comparative analysis of eight different classifiers on the evaluation parameters of accuracy, recall, precision and F-measure. The Random Forest Classifiers offers more accurate predictions than others but LSTM and CNN also give better accuracy.

Underwater wireless sensor network has been an area of interest for a few previous decades. UWSNs consists of tiny sensors responsible for monitoring different underwater events and transmit the collected data to the sink node. In the harsh and continuously changing environment of water, gaining better communication and performance is a difficult task as compared to networks available on land because of different underwater characteristics such as end-to-end delays, node movement, and energy constraints. In this paper, a novel routing technique named angle adjustment for vertical and diagonal communication was proposed, which doesn't use any location information of nodes. It is also efficient in terms of energy and end-to-end delays. In this approach, the source node evaluates the flooding zone based on the angle by using the basic formula for forwarding the packet to the sink. After evaluating the flooding zone, the angles of each node are compared and the packet is sent to the node closest to the vertical line. The proposed approach is evaluated with the help of NS-2 with AquaSim. The results show better results performance in data delivery, end-to-end delays, and energy consumption than DBR.

The analysis of different types of diseases is an extremal vital task which would help in producing vaccines for that particular type of disease. However, this is a very costly process as to test every disease it would mean to analyze every gene related to that specific disease. This issue of genic analysis is further elevated when different variations of each disease is considered. As such the use of different computational methods is taken into consideration to tackle the task of genic variation identification. This research makes use of Machine Learning algorithms to help in the identification and prediction of Single Nucleotide Polymorphism or more specifically Single Amino Acid Polymorphism. Taking into consideration ten different types of diseases, this research makes use of Random Forest and Linear Regression algorithms to identify and predict different genic variations of these diseases. From the extensive research, this article concludes that Random Forest algorithm performs better in comparison to Linear regression in genic variation predictions.