Malaysian Journal of Computer Science

HYBRID DISTANCE-STATISTICAL-BASED PHRASE ALIGNMENT FOR ANALYZING PARALLEL TEXTS IN STANDARD MALAY AND MALAY DIALECTS

Jasmina Khaw Yen Min — 2024-01-31

Parallel texts corpora are essential resources in linguistics and natural language processing, especially in translation and multilingual information retrieval. The publicly available parallel text corpora are limited to certain genres, types and domains. Furthermore, the parallel dialect text is scarce, even though they are important in the analysis and study of a dialect. Collecting parallel dialect text is challenging because dialects typically appear in the form of speech and very limited dialectic texts exist. Moreover, there is no standard orthography in most dialects. The contributions of this paper are threefold. First, the paper describes a methodology in acquiring a parallel text corpus of Standard Malay and Malay dialects, particularly Kelantan Malay and Sarawak Malay. Second, we propose a hybrid of distance based and statistical-based alignment algorithm to align words and phrases the parallel text. The results show that the precision and recall values of the proposed alignment algorithm are more than 95% and better than the state-of the-art GIZA++. Third, the alignment obtained were compared to find out the lexical similarities and differences between Standard Malay and the two studied Malay dialects, contributing valuable insights into the linguistic variations within the Malay language family.

SURVEY ON TECHNICAL ADVANCEMENTS AND RENOVATIONS IN FEDERATED LEARNING

Balaji Subramanian — 2024-01-31

With the rapid increase in IoT devices and advanced machine learning and deep learning techniques, there has been a growing concern about computational cost and data privacy issues since the data coming from IoT devices is non-independent identically distributed (non-IID). However, the implementation of the federated learning algorithm has proven to be a booster in the performance and a solution to the existing data privacy concerns. This paper gives insight into topics such as Blockchain, Unmanned Aerial Vehicles (UAV), Wireless communication, Vehicular Internet of Things, Healthcare, and Cloud Computing and how they have been implemented and co-related to federated Learning and the application and the emerging use cases in the field of federated learning (FL) with respect to the above-mentioned topics have also been discussed. This paper uniquely shows how federated learning has an edge over the traditional machine learning and deep learning techniques in IoT infrastructure since computing nodes are trained using local models on the devices and then these local models are uploaded to the central global server instead of data directly into a global model on a central server ensuring data privacy.

SECURE AND ENERGY-EFFICIENT TASK SCHEDULING IN CLOUD CONTAINER USING VMD-AOA AND ECC-KDF

Muthakshi S — 2024-01-31

Since the cloud storage system offers reliable storage services, it is facing an exponential shift towards lightweight containers due to advancements in technology. Nevertheless, for performing numerous applications, common access to the host Operating System (OS) in the container affected its security. Thus, this paper proposes an energy-efficient and secured scheduling approach. Primarily, multiple users send task requests to the resource manager by registering their details. Then, data are collected from the registered user; also, for removing the repeated requests, pre-processing takes place. Next, by employing the Levenberg-Marquardt Multi-Layer Perceptron Neural Network (LM-MLPNN) technique, the nature of the request from the user is checked to aid in the efficient utilization of container resources. By utilizing the Homography Transform-based K-Mode Algorithm (HT-KMA), the attributes are extracted from the normal user for the efficient clustering process. Then, by deploying the Weighted Round Robin (WRR) technique, the imbalance in the container environment is avoided. An optimal container is selected using the Variational Mode Decomposition-based Archimedes Optimization Algorithm (VMD-AOA) method based on the clustered tasks, and is effectively secured using Elliptic Curve based Key Derivation Function (EC-KDF) and transferred to the resource manager. To perform the required tasks, the resource manager in turn forwards the selected container to the corresponding user. As per the experimental outcomes, when analogized to other well-known algorithms, the proposed methodology achieves better performance.

INVESTIGATING THE IMPORTANCE OF HYPERBOLES TO DETECT SARCASM USING MACHINE LEARNING TECHNIQUES

Vithyatheri Govindan — 2024-01-31

The present study aims to improve sarcasm detection mechanisms using multiple hyperboles such as interjection, intensifiers, capital letters, punctuation, and elongated words. A non-bias dataset consisting of the current pandemic related hashtags was used, namely #Chinesevirus and #Kungflu. Analysis and evaluation were performed with three distinguished machine learning algorithm that is Support Vector Machine, Random Forest and Random Forest with bagging classifiers. Each feature were analysed and the most significant hyperbole identifying sarcasm was assessed further by combining with other hyperboles. The experiments and analysis conducted using these hyperboles concluded that as a single or combined features, hyperboles enhance sarcasm especially in an unbiased dataset.

A NEW APPROACH FOR SPEECH EMOTION RECOGNITION USING SINGLE LAYERED CONVOLUTIONAL NEURAL NETWORK

Mannar Mannan. J — 2024-01-31

Creating a computational device to identify human emotions via voice analysis represents a notable achievement in the sector of human-computer interaction, especially within the healthcare domain. We propose a new light-weight model for addressing challenges of emotions recognition. The model works based on CNN with change of kernel processing. The proposed model performs a direct matching to recognize speech emotions of different eight categories using a statistical model named Analysis of Variance (ANOVA) as kernel for features extraction and Cosine Similarity Measurement (CSM) as activation function for CNN model. This proposed model contains eight-folded single-layered intermediate neurons, and each neuron can segregate speech emotion pattern using CSM from the voice convergence matrix to explore a part of the solution from the whole solution. Experiment results demonstrates that the proposed model outperforms compared with multiple layered existing CNN methods in identifying the emotional state of a speaker.