Journal of Algorithms and Computation
Volume:56 Issue: 1, Aug 2024

In recent years, scholars have dedicated significant attention to the field of sentiment analysis. A substantial volume of feedback shared by tourists on social networking platforms, notably on Tripadvisor, manifests as reviews. The tourism sector stands to gain valuable insights from sentiment analysis applied to such reviews. Despite the extensive body of research in sentiment analysis, scant attention has been directed toward multilingual sentiment analysis, particularly within the domain of tourism. This is noteworthy given the inherently multilingual and global nature of the tourism industry. This study aims to address this gap by presenting a comprehensive multilingual sentiment analysis conducted on Tripadvisor reviews. The sentiment analysis model is crafted using various layers of a neural network. We introduce an augmented Attention-based Bidirectional CNN-RNN Deep Model (Extended ABCDM). Comparative analysis reveals that the multilingual model attains a superior F1 measure of 0.732, outperforming previous models.

This review paper comprehensively examines drone delivery systems, focusing on path planning models, environmental constraints, and application domains. We analyze theoretical frameworks for path planning, including deterministic and heuristic approaches, as well as recent advancements in metaheuristics and hybrid optimization techniques. The paper evaluates how environmental factors, including dynamic obstacles, no-fly zones, and wind conditions, impact drone performance and feasibility. We explore various applications of drone delivery, from last-mile logistics to emergency response, highlighting key challenges and future research directions in this rapidly evolving field. By synthesizing current research and identifying gaps in knowledge, we provide a comprehensive overview to guide future developments in drone delivery systems, with a particular emphasis on recent innovations in multi-objective optimization and adaptive algorithms.

Noise is a part of data whether the data is from measurement or experiment. There are a few techniques for fault detection and reduction to improve the data quality in recent years some of which are based on wavelet, orthogonalization and neural networks. The computational cost of existing methods are more than expected and that's why their application in some cases is not beneficial. In this method, we suggest a tridiagonal model which describes the noise as a function of surrounding signal elements. To make the predicted noise more reliable, the algorithm is equipped with a learning/feedback approach. Our algorithm is used for both small and large noise values. Although the presented numerical results confirm the superlinear convergence of the proposed algorithm, we could only prove the linear convergence. The numerical results confirm the efficiency of presented algorithm in most cases in comparison with orthogonalization based method introduced in 2015.

Let G be a graph. Let f : V (G) → {0, 1, 2,... ,k − 1}be a function where k ∈ N and k > 1. For each edge uv, assign thelabel f (uv) = lf(u)+f(v)2m. f is called a k-total mean cordial labeling of G if |tmf (i) − tmf (j)| ≤ 1, for all i,j ∈ {0, 1, 2,... ,k − 1},where tmf (x) denotes the total number of vertices and edges labelled with x, x ∈ {0, 1, 2,... ,k − 1}. A graph with admit a k-totalmean cordial labeling is called k-total mean cordial graph. In thispaper we examine the 4-Total mean cordial labeling of some trees

Persian digit recognition plays a crucial role in computer vision and pattern recognition. Existing algorithms fall into two categories: traditional methods and deep learning approaches. While many deep learning techniques are documented, they often depend on pre-trained networks with numerous parameters, requiring substantial resources and time for training and prediction. This paper presents a novel convolutional neural network (CNN) architecture for Persian digit recognition that is shallower than current models, thereby reducing the number of trainable parameters. We introduce dilated convolution layers to capture larger features without increasing parameters and propose a combined loss function to improve accuracy. Trained on the HODA dataset, our method achieves a validation accuracy of 99.82\%, test accuracy of 99.79\%, and training accuracy of 100\%. The proposed network demonstrates enhanced accuracy, faster performance, and significantly reduced implementation time due to its streamlined architecture.

In this paper, a three-stage CMOS inverter based ring voltage controlled oscillator (VCO) is designed. Low power consumption is an important feature of the designed VCO. The variables used in the designed oscillator have been quantified using the improved Gravitational Search Algorithm (GSA) in order to minimize the power consumption of the oscillator. In this algorithm, the Boltzmann scaling function is used to control the exploration and productivity capabilities of GSA. Other advantages of the designed VCO are the high integration capability, simplicity of implementation and high frequency tuning range. The power dissipation of the VCO is 850 µW. The presented oscillator can operate from 1.25 GHz to 2.5 GHz frequency.

This paper is about metric and partition dimension of a flower and a pencilgraph. A metric dimension of G, denoted by dim(G), is the minimum cardinality of anyresolving set of G. A partition dimension of G, denoted by pd(G), is the minimum number of sets in any resolving k-ordered partition for G. Here we give the exact value of themetric dimension of a flower graph fm×n for m ∈ {3, 4} and a pencil graph Pcm for anyinteger m ≥ 2. We also give the partition dimension of fm×n for m ∈ {3, 4, 5} and Pcmfor any integer m ≥ 2.

Web recommender systems provide the most appropriate recommendations by analyzing user’s navigation behavior. This recommender system can be considered in different cases, such as e-commerce, search engines, etc. The aim of the proposed approach in this research is to create users’ profiles and find their common navigation patterns implicitly. The web log file is utilized to analyze browsing history and discover users' navigation models. This analysis is called web usage mining. This research focused on the K-means algorithm as a cluster, and the neural network as a classification algorithm, along with the recommended Markov model. The innovation of this research is to consider a threshold for the proposed Markov model. The main goal of this research is to create a recommender system based on the Markov model and neural network that provides an acceptable suggestion with high accuracy and precision.

The paper introduces a new method called ABCL-EHI for human identification using electroencephalographic (EEG) signals. EEG signals have unique information among individuals, but current systems lack accuracy and usability. ABCL-EHI addresses this by combining a convolutional neural network (CNN) and a long short-term memory (LSTM) network with an attention mechanism. This attention mechanism enhances the utilization of spatial and temporal characteristics of EEG signals. The proposed system is evaluated using a public dataset of EEG signals from 109 subjects performing motor/imagery tasks. The results demonstrate that ABCL-EHI achieves high accuracy, with F1-Score scores of 99.65, 99.65, and 99.52 when using 64, 14, and 9 EEG channels, respectively. This outperforms previous studies and highlights the system's reliability and ease of deployment in real-life applications, as it maintains high accuracy even with a small number of EEG channels and allows users to perform various tasks while recording signals.

Cloud computing allows users to access software and hardware resources over the network. To achieve cost-effective execution, task scheduling is a major problem. Scheduling tasks is intended to optimize one or more criteria by allocating them to resources. Metaheuristics provide promising solutions for task scheduling by leveraging bio-inspired techniques. In order to solve the task scheduling problem, we present an algorithm that uses a new meta-heuristic algorithm namely WO (Walrus optimizer). The proposed method is known as WOTSA (WO-based Task Scheduling Algorithm). Its main objective is to decrease execution cost, load balancing, resource utilization, and makespan. WOTSA is compared against several popular meta-heuristic algorithms, including FOX (Fox Optimizer), GEO (Golden Eagle Optimizer), ZOA (Zebra Optimization Algorithm), and STOA (Sooty Tern Optimization Algorithm). According to the experimental results, WOTSA improves performance in terms of makespan, resource utilization, execution cost, and degree of resource load balance.