Search

The accurate prediction of demographic attributes of users, including age and gender, is a pivotal challenge in personalized search, ad targeting, and other related fields. This information enables companies to refine their target audience and enhance the overall user experience and quality of service (QoS). Among these, the Saudi Telecommunication Company (STC), a premier telecommunications provider in Saudi Arabia, Middle East, and Africa, recognizes the substantial role of age-prediction systems. This study, therefore, explores the application of machine learning (ML) techniques to predict user age, thus assisting in the delivery of age-appropriate ads and offers. We utilized a dataset provided by STC, comprising three million samples with key user and device features. Four ML algorithms were employed in this analysis: the Artificial Neural Network (ANN), Random Forest (RF), Gradient Boosting (GB), and Decision Tree (DT). These models were compared and evaluated based on their predictive performance. The ANN emerged as the optimal classifier, achieving an accuracy of 60%, comparable to similar studies conducted within the telecommunications industry. The implications of these findings suggest that ML techniques can effectively predict user information, thereby enabling service providers to tailor their offerings to the specific age demographics of their users. The findings from this study contribute to the broader understanding of user age prediction and its practical implications for telecommunications companies. Future research could extend this work by exploring other demographic prediction challenges and applying the ML approach to other sectors. [ABSTRACT FROM AUTHOR]

With the rapid advances in Information Technology (IT), the focus on engaging computing students to gain practical experience in the IT industry before graduation is becoming increasingly complex without incorporating pedagogical strategies of success in curricula. The goal is to enable computing major students to gain in-depth knowledge and practical understanding of the IT working environment before graduating through essential industry-driven practical skills based on international standards and best practices. Unfortunately, tracking and analyzing students' practical skills performance during their IT field training programs, which are conducted primarily off-campus at various public and private organizations, before, during, and after the training period, is a daunting task for both the college instructors and the industry trainers. To overcome these challenges, this paper introduces a Sustainable University Field Training (SUNFIT) framework, which is a pedagogical approach towards mining the educational data using machine learning to integrate and measure the field training programs against the internationally recognized accreditation standards such as Accreditation Board for Engineering and Technology (ABET). The study employs machine learning models aimed at continuously measuring and monitoring international ABET accreditation requirements on computing major courses' academic data, elucidating student performance across various semesters, integrating best practices, and producing an evidence-based rationale approach for evaluating weak learning outcomes (LOs) with minimal manual intervention, as well as preventing faculty-specific portfolio errors. The proposed approach could be easily developed by academics, researchers, or even students, and for a variety of purposes, including enhancing poor student outcomes (SOs). In addition, various data mining and machine learning approaches have been investigated over field training assessment data for successful prediction in subsequent cycles. The results are promising, with Naïve Bayes obtaining the highest accuracy of 90.54% followed by J48 and PART algorithms at 87.83%. [ABSTRACT FROM AUTHOR]

With the emergence of the Internet of Things (IoT), investigation of different diseases in healthcare improved, and cloud computing helped to centralize the data and to access patient records throughout the world. In this way, the electrocardiogram (ECG) is used to diagnose heart diseases or abnormalities. The machine learning techniques have been used previously but are feature-based and not as accurate as transfer learning; the proposed development and validation of embedded device prove ECG arrhythmia by using the transfer learning (DVEEA-TL) model. This model is the combination of hardware, software, and two datasets that are augmented and fused and further finds the accuracy results in high proportion as compared to the previous work and research. In the proposed model, a new dataset is made by the combination of the Kaggle dataset and the other, which is made by taking the real-time healthy and unhealthy datasets, and later, the AlexNet transfer learning approach is applied to get a more accurate reading in terms of ECG signals. In this proposed research, the DVEEA-TL model diagnoses the heart abnormality in respect of accuracy during the training and validation stages as 99.9% and 99.8%, respectively, which is the best and more reliable approach as compared to the previous research in this field. [ABSTRACT FROM AUTHOR]

Air pollution is a significant problem in modern societies since it has a serious impact on human health and the environment. Particulate Matter (PM2.5) is a type of air pollution that contains of interrupted elements with a diameter less than or equal to 2.5 m. For risk assessment and epidemiological investigations, a better knowledge of the spatiotemporal variation of PM2.5 concentration in a constant space-time area is essential. Conventional spatiotemporal interpolation approaches commonly relying on robust presumption by limiting interpolation algorithms to those with explicit and basic mathematical expression, ignoring a plethora of hidden but crucial manipulating aspects. Many advanced deep learning approaches have been proposed to forecast Particulate Matter (PM2.5). Recurrent neural network (RNN) is one of the popular deep learning architectures which is widely employed in PM2.5 concentration forecasting. In this research, we proposed a Two-State Gated Recurrent Unit (TS-GRU) for monitoring and estimating the PM2.5 concentration forecasting system. The proposed algorithm is capable of considering both spatial and temporal hidden affecting elements spontaneously. We tested our model using data from daily PM2.5 dimensions taken in the contactual southeast area of the United States in 2009. In the studies, three evaluation matrices were utilized to compare the overall performance of each algorithm: Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The experimental results revealed that our proposed TS-GRU model outperformed compared to the other deep learning approaches in terms of forecasting performance. [ABSTRACT FROM AUTHOR]

This paper is the extension of the paper stated in [1].In the extended paper, elaboration of outlier entities are discovered based on the behavioral dissimilarities. Fifteen different features were proposed for this process using a variant of Kth nearest neighborhood (KNN) algorithm. Outliers were the convicted email users in ENRON, US based company which were already been declared convicted. The results in the papers proved to be matched with the 80% of the convicted email users because few of the users were not detected due to the constraints of the algorithm. In top 19 outliers detection 3 convicted users were found out. It means 15% of the result was achieved in top 20 users. The bench marking is made with the existing convicted and declared email users. The proposed features proved to helpful in the detection of outliers in email communication network and can be further implemented for various other kind of communication networks. [ABSTRACT FROM AUTHOR]

The Biometric Gaussian Stream (BGS) cryptosystem was extended by extensive research experimentation. Using this' system, complexity is added to an image by passing it through a Gaussian noise function. This function is applied with specific parameters for the mean and variance, which also works as a parallel key. To implement a stream cipher BGS with help of biometric images, the Initial Condition (IC) for Linear Feed Back Shift Register (LFSR) from is extracted iriscode. A comparison between various stream ciphers is also made to measure the strength of the BGS cryptosystem. The previous experimentation work has been extended by formulating the algorithmic runtime complexity of the BGS Cryptosystem which proves to be O(n) algorithmically. New techniques to encrypt and assess multimedia objects have been introduced. [ABSTRACT FROM AUTHOR]

A novel reversible digital watermarking technique for medical images to achieve high level of secrecy, tamper detection, and blind recovery of the original image is proposed. The technique selects some of the pixels from the host image using chaotic key for embedding a chaotically generated watermark. The rest of the pixels are converted to residues by using the Residue Number System (RNS). The chaotically selected pixels are represented by the polynomial. A primitive polynomial of degree four is chosen that divides the message polynomial and consequently the remainder is obtained. The obtained remainder is XORed with the watermark and appended along with the message. The decoder receives the appended message and divides it by the same primitive polynomial and calculates the remainder. The authenticity of watermark is done based on the remainder that is valid, if it is zero and invalid otherwise. On the other hand, residue is divided with a primitive polynomial of degree 3 and the obtained remainder is appended with residue. The secrecy of proposed system is considerably high. It will be almost impossible for the intruder to find out which pixels are watermarked and which are just residue. Moreover, the proposed system also ensures high security due to four keys used in chaotic map. Effectiveness of the scheme is validated through MATLAB simulations and comparison with a similar technique. [ABSTRACT FROM AUTHOR]