In order to solve the problem that the service composition in the logistics field has poor tolerance and unreliable service, a model of logistics service fault-tolerant composition for intelligent logistics system of Internet of Things (IoT) based on π-net was built. Firstly, after a brief introdution of IoT intelligent logistics system, a fault-tolerant service composition framework for the system was provided. Then, a model of logistics service fault-tolerant composition for the system based on π-net was built, and the correctness of fault tolerance and fitting degree of the model were analyzed. Finally, the service reliability and the fault-tolerant reliability of the model were tested, and the comparison with Petri-net, QoS (Quality of Service) dynamic prediction, fuzzy Kano model and modified particle swarm optimization methods in the service composition execution time, user satisfaction, reliability and optimal degree were carried out. The results show that the proposed model has high service reliability and fault-tolerant reliability, and has certain advantages in terms of service composition execution time, user satisfaction, reliability and optimal degree.

Multi-Intention (MI) recognition of short texts is a problem in Spoken Language Understanding (SLU). The effective features of short texts are difficult to extract in classification problems because of sparse features of short texts and few words containing many information. To solve the problem, by combining syntactic features and Convolution Neural Network (CNN), a multi-intention recognition model was proposed. Firstly, the sentence was syntactically analyzed to determine whether it contains multi-intention. Secondly, the number of intentions and matrix of distance were calculated by using Term Frequency-Inverse Document Frequency (TF-IDF) and word embedding. Then the matrix of distance was acted as the input of CNN model to classify intentions. Finally, the emotional polarity of each intention was judged to return to the user's true intentions. The experiment was carried out by using the real data of the existing intelligent customer service system. The experimental results show that, the single classification precision of the combination model of syntactic features and CNN is 93.5% in 10 intentions, which is 1.4 percentage points higher than the original CNN model without syntactic features. And in mutil-intention recognition, the classification precision is 30 percentage points higher than others.

Aiming at the problems of slow transcoding speed of single-machine video transcoding method and limited efficiency improvement of parallel transcoding method for batch processing, a fast video transcoding method for stream processing based on Spark Streaming distributed stream processing framework was proposed. Firstly, an automated video slicing model was built by using the open source multimedia processing tool of FFmpeg and a programming algorithm was proposed. Then, in view of the characteristics of parallel video transcoding, the stream processing model of video transcoding was constructed by studying Resilient Distributed Datasets (RDD). Finally, the video merging scheme was designed to store the combined video files effectively. Based on the proposed fast video transcoding method, a fast video transcoding system based on Spark Streaming was designed and implemented. The experimental results show that, compared with the Hadoop video transcoding method for batch processing, the proposed method has improved the transcoding efficiency by 26.7%, and compared with the video parallel transcoding based on Hadoop platform, the proposed method has improved the transcoding efficiency by 20.1%.

Aimming at the problem of leaking privacy to Third-Party Auditors (TPA) and initiating alternative attacks by Cloud Storage Server (CSS) in public auditing, a new public auditing scheme of data integrity for public cloud was proposed. Firstly, the hash value obfuscation method was used to obfuscate the evidence returned by the cloud storage server to prevent TPA from analyzing and calculating the original data. Then during the audit process, TPA itself calculated the overlay tree of the Merkle Hash Tree (MHT) corresponding to the challenge request, and matched with the overlay tree returned by CSS to prevent the cloud storage server from responding to audit challenges with other existing data. Experimental results show that the performance in terms of computational overhead, storage overhead and communication overhead does not change by orders of magnitude after solving the privacy and attack problems of the existing scheme.