Behavioral Targeting (BT) is a recent trend of online advertising market. However, some classical BT solutions, which predefine the user segments for BT ads delivery, are sometimes too large to numerous long-tail advertisers, who cannot afford to buy any large user segments due to budget consideration. In this extend abstract, we propose to rank users according to their probability of interest in an advertisement in a learning to rank framework. We propose to extract three types of features between user behaviors such as search queries, ad click history etc and the ad content provided by advertisers. Through this way, a long-tail advertiser can select a certain number of top ranked users as needed from the user segments for ads delivery. In the experiments, we use a 30-days' ad click-through log from a commercial search engine. The results show that using our proposed features under a learning to rank framework, we can well rank users who potentially interest in an advertisement.

Wireless sensor networks are often deployed in harsh environments, where the adversary is able to capture certain sensors. Once a sensor is compromised, the adversary can easily replicate it and deploy several replicas back into the network for further malicious activities. Although a number of protocols have been proposed to tackle such node replication attacks, few of these schemes are suitable for mobile wireless sensor networks. In this paper, we propose two novel mobility-assisted distributed solutions to node replication detection in mobile wireless sensor networks. In both protocols, after receiving the time-location claims, witnesses carry these claims around the network instead of transmitting them. That means data are forwarded only when appropriate witnesses encounter each other. Unary-Time-Location Storage & Exchange (UTLSE) detects the replicas by each of the two encountered witnesses which stores only one time-location claim. Multi-Time-Location Storage & Diffusion (MTLSD), by storing more time-location claims for each tracked node and introducing time-location claims diffusion among witnesses, provides excellent resiliency and sub-optimal detection probability with modest communication overhead. Due to the mobility-assisted property, our protocols do not rely on any specific routing protocol, which makes them suitable for various mobile settings. Our theoretical analysis and simulation results show that our protocols are efficient in terms of detection performance, communication overhead and storage overhead.