The increase in the amount of data stored in the cloud leads to the need for privacy-preserving multi-keyword search schemes in the cloud. However, most of the existing schemes usually adopt the TF-IDF vector space model, in which the vectors are high-dimensional and sparse. It results in substantial computation time and storage space. To address the issue, we propose an efficient dictionary partition-based multi-keyword ranked search scheme (DPMRS) over encrypted cloud data. First, a dictionary partition-based vector space model (DPVSM) is designed, which can compress vector dimensions and hence accelerate relevance score computation between documents and search keywords. Based on DPVSM, a dictionary partition-based keyword distribution inverted index (DPKD-index) is presented. By using the index, a baseline privacy-preserving ranked search scheme is proposed. To further improve the efficiency of search services, the search tree structure is adopted and a novel double tier search tree-based index (DSTree-index) is designed. By using the optimized index, an enhanced search scheme (DPMRS+) is proposed. The security analysis indicates that the proposed scheme can protect the privacy of search processing, and the experimental results show that the proposed scheme outperforms the existing works in terms of storage size, search precision, and search time cost.

This research is dedicated to addressing user recommendation matching and multi-objective optimization problems in ride-sharing services. For addressing the challenge of node classification in social networks, the Graph Attention Network with Opinion Dynamics (OD-GAT) is proposed. This model combines opinion dynamics and attention mechanism, which can make full use of multi-dimensional information for social relationship reasoning, and at the same time simulate the influence of individuals by other objects in the group, realize more accurate prediction and reasoning of social relationships, improved service quality and ride-sharing safety. To address the intricate task of balancing multiple objectives, including average detour cost, average response rate, and average user similarity rate, we introduce a novel evolutionary computation method for optimizing ride-sharing scenarios. This approach tackles the dynamic ride-sharing matching problem by emphasizing human factors in the optimization goals, successfully overcoming challenges related to local optima and convergence. Experimental validation confirms the effectiveness of OD-GAT in feature extraction and classification, showcasing the method’s fastest convergence speed and global optimum achievement across three key metrics.

This study addresses the Multiple Flying Sidekicks Traveling Salesman Problem (mFSTSP), where parallel Unmanned Aerial Vehicles (UAVs, or Drones) work alongside truck to enhance delivery efficiency. Existing scheduling approaches face challenges in high computational costs and the risk of converging to local optima due to excessive exploration in unknown environments, especially in large-scale mFSTSP. This study proposed a Large Language Model Enhanced Q-Learning Approach (LLM-QL) to solve mFSTSP, which combines the local exploration advantages of Q-Learning with the global understanding of unknown environments provided by LLMs, thus improving the efficiency of path planning. A novel prompt strategy is also provided, transforming the problem modeling into a format easily understood by LLMs, guiding the algorithm's exploration and significantly improving convergence. We also provide a proof of the convergence of LLM-QL. Experimental results demonstrate that LLM-QL achieves up to a 1.35 x improvement in key performance metrics such as total completion time, algorithm runtime, and UAV utilization, compared to existing state-of-the-art methods.

Recent advancement in propulsion technologies to reduce the need for travel or increase the share of sustainable unmanned devices has accelerated the shift toward sustainable transport. To achieve the optimization of route planning in the air-ground integrated network (AGIN), we design an optimization strategy of accompanying graph navigation for unmanned devices, which aims to reduce the power consumption and$CO_{2}$gas emissions. The optimization of accompanying graph navigation is composed of three strategies, namely, the navigation based on the complete maps, the navigation based on the partitioned maps, and the navigation without maps. We propose a Two-tiered Grid (TG) index and Distributed AGIN Navigation (DAN) to navigate on partitioned maps. The top layer of the TG-index is composed of the border vertices of the global road network, which reflects the overall traffic conditions of the global road network and provides coarse-grained navigation routes. The bottom layer is a grid index composed of subgraphs, which reflects traffic conditions in local areas and provides fine-grained navigation routes. The navigation optimization is implemented in several segments, which can be run by multi-processors and realize rapid response to a large number of concurrent queries.

The document vectors constructed by the traditional searchable encryption scheme based on the term frequency-inverse document frequency model not only have high dimensionality and sparsity, but also ignore the semantic information of documents and keywords. In this paper, we introduce the sentence bidirectional encoder representations from transformers model (SBERT) to obtain semantic information-embedded vectors for documents and keywords. By adopting the SBERT model, we pro-pose a CBG-index based semantic-aware multi-keyword ranked search scheme (CSMRS). In the scheme, a topic-term frequency-inverse topic frequency (TTF-ITF) model and a clustering-based group index (CBG-index) are proposed. The TTF-ITF model is used to generate semantic vectors for keywords, and the CBG-index is used to improve the search efficiency. The experimental results demonstrate the better performance than the existing works in terms of search efficiency and search result semantic precision.

Traditional term frequency-inverse document frequency model-based privacy-preserving ranked search schemes rarely consider the latent semantic meanings of documents and keywords. It is a challenge to design efficient semantic-aware ranked search (SRSE) schemes with privacy preservation. In this paper, two privacy-preserving SRSE schemes are developed for the cloud environments. The first scheme is the accuracy-first search scheme. In this scheme, the Latent Dirichlet Allocation topic model is adopted to generate the topic-based semantic information-embedded vectors for documents and queried keywords, which supports semantic-aware relevance measurement. The bisecting k-means clustering algorithm is used to build an accuracy-first filtering tree index (AFF-tree), and the AFF-tree-based search algorithm is proposed to achieve the accuracy-first ranked search. The second scheme is the efficiency-first search scheme. It performs a structure optimization on the AFF-tree, and a newly efficiency-first filtering tree index (EFF-tree) is designed. By using the EFF-tree, an anchor node-based search algorithm is designed to achieve the efficiency-first ranked search at the expense of a little decrease in search result precision. The secure inner product is used to perform privacy-preserving semantic-aware relevance measurement between documents and queried keywords in both schemes. To analyze the security of the proposed schemes, the game stimulation-based proof is presented. Experimental results show the better performance of the proposed schemes in search time cost.