Biblio

Research on advertisement has mainly focused on how to accurately predict the click-through rate (CTR). Much less is known about fraud detection and malicious behavior defense. Previous studies usually use statistics, design threshold and manually make strategies, which cannot find potential fraud behavior effectively and suffer from new attacks. In this paper, we make the first step to understand the type of malicious activities on large-scale online advertising platforms. By analyzing each feature comprehensively, we propose a novel coding approach to transform nominal attributes into numeric while maintaining the most effective information of the original data for fraud detection. Next, we code important features such as IP and cookie in our dataset and train machine learning methods to detect fraud traffic automatically. Experimental results on real datasets demonstrate that the proposed fraud detection method performs well considering both the accuracy and efficiency. Finally, we conclude how to design a defense system by considering which methods could be used for the anti-spam gaming in the future.

Providing recommendations on social systems has been in the spotlight of both academics and industry for some time already. Social network giants like Facebook, LinkedIn, Myspace, etc., are eager to find the silver bullet of recommendation. These applications permit clients to shape a few certain social networks through their day-by-day social cooperative communications. In the meantime, today's online experience depends progressively on social association. One of the main concerns in social network is establishing a successful business plan to make more profit from the social network. Doing a business on every platform needs a good business plan with some important solutions such as advertise the products or services of other companies which would be a kind of marketing for those external businesses. In this study a philosophy of a system speaking to of a comprehensive structure of advertisement recommender system for social networks will be presented. The framework uses a semantic logic to provide the recommended products and this capability can differentiate the recommender part of the framework from classical recommender methods. Briefly, the framework proposed in this study has been designed in a form that can generate advertisement recommendations in a simplified and effective way for social network users.