A hashtag based method was proposed to solve the problem how to accurately extract keywords from microblog. Hashtag, the social feature of a microblog was used to extract keywords from microblog content. A word-post weighted graph was built firstly, then a random walker was used on the graph by jumping to any hashtag node repeatedly. At last, every word rank was determined by its probability which would not change after walker iteration. The experiments were conducted on real microblogs from Sina platform. The results show that, compared to word-word graph method, the proposed hashtag-based approach gets higher accuracy of keyword extraction by 50%.

Most of the variants of Graph Cut algorithm do not impose any shape constraints on the segmentations, rendering it difficult to obtain semantic valid segmentation results. As for pedestrian segmentation, this difficulty leads to the non-human shape of the segmented object. An improved Graph Cut algorithm combining shape priors and discriminatively learned appearance model was proposed in this paper to segment pedestrians in static images. In this approach, a large number of real pedestrian silhouettes were used to encode the a'priori shape of pedestrians, and a hierarchical model of pedestrian template was built to reduce the matching time, which would hopefully bias the segmentation results to be humanlike. A discriminative appearance model of the pedestrian was also proposed in this paper to better distinguish persons from the background. The experimental results verify the improved performance of this approach.

For reducing the communication overhead of traditional data aggregation method in wireless sensor networks, the authors proposed a probabilistic transmissionbased data aggregation scheme for Wireless Sensor Network (WSN). Due to limited number of nodes in the cluster and the fact that aggregation error is unavoidable, probabilistic transmission was adopted to reduce the number of innercluster transmissions and lower the communication overhead with tolerable error. Besides, Dixon criterion was adopted to eliminate the gross error in the small sample to provide high reliability of innercluster aggregation. The experimental results show that the probabilistic transmission can lower the innercluster transmissions effectively with tolerable error, the communication overhead of proposed scheme is about 27.5% that of traditional data aggregation schemes. The aggregation error of probabilistic transmission and all nodes transmission are at the same level and both are acceptable for wireless sensor networks.

Feature dependency modeling describes the model of mutual constraints among features. It is an essential activity for development of software product line. This paper introduced local dependency and global dependency of feature. With analyzing relationship of feature dependency, an approach of feature dependency modeling was proposed. It not only provides specification of local dependency such as decomposition and generalization, but also provides global dependency modeling such as configuration dependency, operational dependency and impact dependency. Finally a case study of feature dependency modeling in air-condition controlling software product line is presented to demonstrate our method.

This paper introduced UML to software product line. With analyzing variability of product line, we propose an approach of variability modeling for software product line with UML. It not only provides specification of variability type such as option and alternative, but also provides constrains modeling of software product line variability. Finally a case study of variability modeling in mobile software product line was presented to demonstrate our method.

Because the heterogeneous information is relatively loose and independent in seamless transportation, data grid is adopted to manage the distributed heterogeneous information and realize resources share and information integration. The paper presents a architecture of seamless transportation information grid(STIG),the architecture integrates Information integration, semantic query, transportation planning, etc key technologies. At last, take combined rail and sea transport for example, the process of multimodal transportation based on STIG was described.

The key to body movement emotion recognition lies in extracting emotional features existed in human body movements. To solve the problems of poor emotional feature learning capability and difficulty in improving emotion recognition accuracy in existing models， a body movement emotion recognition method based on Emotional Latent Space Learning （ELSL） and Contrastive Language-Image Pre-training （CLIP） model was proposed. Firstly， CLIP model was introduced to improve the emotional feature learning capability of the model. Secondly， for the fine-grained multi-label emotion classification task， ELSL method was proposed. By learning discriminative mappings from emotional latent space to various subspaces， the subtle differences between emotion categories and the feature information beneficial to the classification of each emotion category in various emotional subspaces. Experiments were carried out on real-world open scenarios-oriented Body Language Dataset （BoLD） The results demonstrate that the proposed method makes use of the advantages of CLIP model and latent space learning in feature learning effectively， leading to significant performance improvement. In specific， compared to Movement Analysis Network （MANet）， the proposed method has a 1.08 percentage points increase in mean Average Precision （mAP） and a 1.32 percentage points improvement in mean Area Under Receiver Operating Characteristic Curve （mRA）.