The advertising ecosystem faces major threats from ad fraud caused by artificial display requests or clicks, created by malicious codes, bot-nets, and click-firms. Currently, there is a multibillion-dollar online advertisement market which generates the primary revenue for some of the internet's most successful websites. Unfortunately, the complexities of the advertisement ecosystem attract a considerable amount of cybercrime activity, which profits at the expense of advertisers. Web ad fraud has been extensively studied whereas fraud in mobile ads has received very little attention. Most of these studies have been carried out to identify fraudulent online and mobile ads clicks. However, the identification of individual fraudulent displays in mobile ads has yet to be explored. Additionally, other fraudulent activity aspects such as hacking ad-campaign accounts have rarely been addressed. The purpose of this study is to provide a comprehensive review of state-of-the-art ad fraud in web content as well as mobile apps. In this context, we will introduce a deeper understanding of vulnerabilities of online/mobile advertising ecosystems, the ad fraud’s well-known attacks, their effective detection methods and prevention mechanisms. arabic 8 English 40
Hala Shaari, Nuredin Ahmed(12-2020)

Machine learning is the science of making machines work without predefined rules being programmed . You typically provide the training data with features in machine learning, such as the type of words with output variables such as spam or ham. Once that data is fed into learning algorithms, such as classification or regression, it learns a model of correlation between features and output variables. You can predict that the e-mail is a spam or ham by providing input e-mails called test data to the model. By providing more and more training data you can refine the model to improve accuracy
Hala Shaari, Jasmin Kevrić, Samed Jukić (10-2022)

Electricity is one of the fundamental necessities of human beings, which has many uses in our day to day life. It is used for different purposes like domestic, industrial and agricultural. The biggest challenge facing electricity distribution is data collection and meter reading. Right now, meter reading is collected manually which give scope for corruption and human error in reading, moreover the wastage of manpower and resources of utility. Prepaid Energy Meter has been implemented in several countries. In fact, the disadvantage of the system is the behavioral control of the users. Moreover, recharging should be carried out on the meter. The problem occurs when consumers leave their premises and electrical pulses are discharged. That's why we need a system to control the electrical pulse wherever they are. In this work, a prepaid energy meter was proposed, implemented and simulated using PROTEUS software. The system was designed using ATmega128 as a microcontroller and GSM technology is advancement over conventional energy meter, which enables consumer to effectively manage their electricity usage. Additionally, it evaluates the accuracy of voltage and current measured by means of this model. Our Suggested model of the prepaid power meter produces the lowest error compared to actual voltage and current. The proposed system replaces traditional meter reading methods and enables remote monitor and control the meter readings regularly not manually. Also, it alerts the consumer when the energy consumption exceeds above the set limit and alerts the utility company if there is any theft that might be happened.
Nuredin Ali Salem Ahmed, Hala Shaari, Adel Saad Emhemmed(5-2021)