Code vulnerability detection is critical for software security prevention. Vulnerability annotation in large-scale software code is quite tedious and challenging, which requires domain experts to spend a lot of time annotating. This work offers CPVD, a cross-domain vulnerability detection approach based on the challenge of ”learning to predict the vulnerability labels of another item quickly using one item with rich vulnerability labels.” CPVD uses the code property graph to represent the code and uses the Graph Attention Network and Convolution Pooling Network to extract the graph feature vector. It reduces the distribution between the source domain and target domain data in the Domain Adaptation Representation Learning stage for cross-domain vulnerability detection. In this paper, we test each other on different real-world project codes. Compared with methods without domain adaptation and domain adaptation methods based on natural language processing, CPVD is more general and performs better in cross-domain vulnerability detection tasks. Specifically, for the four datasets of chr_deb, qemu, libav, and sard, they achieved the best results of 70.2%, 81.1%, 59.7%, and 78.1% respectively on the F1-Score, and 88.4%,86.3%, 85.2%, and 88.6% on the AUC.

With the development of mobile applications, location-based services (LBSs) have been incorporated into people’s daily lives and created huge commercial revenues. However, when using these services, people also face the risk of personal privacy breaches due to the release of location and query content. Many existing location privacy protection schemes with centralized architectures assume that anonymous servers are secure and trustworthy. This assumption is difficult to guarantee in real applications. To solve the problem of relying on the security and trustworthiness of anonymous servers, we propose a Geohash-based location privacy protection scheme for snapshot queries. It is named $GLPS$. On the user side, $GLPS$ uses Geohash encoding technology to convert the user’s location coordinates into a string code representing a rectangular geographic area. $GLPS$ uses the code as the privacy location to send check-ins and queries to the anonymous server and to avoid the anonymous server gaining the user’s exact location. On the anonymous server side, the scheme takes advantage of Geohash codes’ geospatial gridding capabilities and GL-Tree’s effective location retrieval performance to generate a k-anonymous query set based on user-defined minimum and maximum hidden cells, making it harder for adversaries to pinpoint the user’s location. We experimentally tested the performance of $GLPS$ and compared it with three schemes: $Casper$, $GCasper$, and $DLS$. The experimental results and analyses demonstrate that $GLPS$ has a good performance and privacy protection capability, which resolves the reliance on the security and trustworthiness of anonymous servers. It also resists attacks involving background knowledge, regional centers, homogenization, distribution density, and identity association.

Location-based application services and location privacy protection solutions are often required for the storage, management, and efficient retrieval of large amounts of geolocation data for specific locations or location intervals. We design a hierarchical tree-like organization structure, GL-Tree, which enables the storage, management, and retrieval of massive location data and satisfies the user’s location-hiding requirements. We first use Geohash encoding to convert the two-dimensional geospatial coordinates of locations into one-dimensional strings and construct the GL-Tree based on the Geohash encoding principle. We gradually reduce the location intervals by extending the length of the Geohash code to achieve geospatial grid division and spatial approximation of user locations. The hierarchical tree structure of GL-Tree reflects the correspondence between Geohash codes and geographic intervals. Users and their location relationships are recorded in the leaf nodes at each level of the hierarchical GL-Tree. In top–down order, along the GL-Tree, efficient storage and retrieval of location sets for specified locations and specified intervals can be achieved. We conducted experimental tests on the Gowalla public dataset and compared the performance of the B+ tree, R tree, and GL-Tree in terms of time consumption in three aspects: tree construction, location insertion, and location retrieval, and the results show that GL-Tree has good performance in terms of time consumption.