In recent years, research in software security has concentrated on identifying vulnerabilities in smart contracts to prevent significant losses of crypto assets on blockchains. Despite early successes in this area, detecting developers' intents in smart contracts has become a more pressing issue, as malicious intents have caused substantial financial losses. Unfortunately, existing research lacks effective methods for detecting development intents in smart contracts. To address this gap, we propose SMARTINTENTNN (Smart Contract Intent Neural Network), a deep learning model designed to automatically detect development intents in smart contracts. SMARTINTENTNN leverages a pre-trained sentence encoder to generate contextual representations of smart contracts, employs a K-means clustering model to identify and highlight prominent intent features, and utilizes a bidirectional LSTM-based deep neural network for multi-label classification. We trained and evaluated SMARTINTENTNN on a dataset containing over 40,000 real-world smart contracts, employing self-comparison baselines in our experimental setup. The results show that SMARTINTENTNN achieves an F1-score of 0.8633 in identifying intents across 10 distinct categories, outperforming all baselines and addressing the gap in smart contract detection by incorporating intent analysis.

As an essential branch of web service applications, the location-based service (LBS) plays an irreplaceable role in our daily lives. Usually, the LBS is time-sensitive, which requires the system to process trajectory data in a real-time manner. Due to the sensitivity of trajectory data, LBS services may violate users’ privacy. Furthermore, trajectory compression plays a crucial auxiliary role in analyzing and mining massive trajectory raw data such as trajectory clustering and trajectory similarity calculation and can help keep users’ privacy. In other words, trajectory compression serves as the prerequisite for privacy-preserved trajectory data mining, which retains points with high-information content and removes redundant approximate points with low information value under the premise of protecting users’ privacy. We can speed up the applications’ response speed and save computing resources if we take advantage of trajectory compression and provide lightweight data support for big-data-driven web page extraction, convenient for fast and accurate response. Unfortunately, trajectory compression’s current real-time processing capacity is still not big enough and not cost-effective. In terms of the implementation principle, most of the existing works are micro-batch processing. Consequently, the system will overly consume resources and respond with a high latency with the trajectory data inputting. In addition, it is difficult for users to understand and set compression parameters correctly. In this context, we propose an algorithm to incrementally compress the trajectory in real-time based on the azimuth change, and two kinds of user-perceivable parameters are proposed to facilitate real-time specific compression. For verification, our study uses real-world data sets, such as GeoLife Trajectory data. We also found that compared with the current OPW-TR algorithm with better all-around performance, our algorithm dramatically improves the processing speed with a minimal loss of accuracy. Furthermore, thanks to the maintenance of incremental stateful computation, our memory consumption was reduced by 28.5% when processing about 400k records. The memory advantage will become more pronounced as the number of data increases.