With the rapid development of technologies such as modern network communication and social media, the networked big data is difficult to be applied due to the lack of efficient and available node representation. Network representation learning is widely concerned by transforming high-dimensional sparse network data into low-dimensional, compact and easy-to-apply node representation. However, the existing network embedding methods obtain the low-dimensional feature vectors of nodes and then use them as the inputs for other applications (such as node classification, community discovery, link prediction and visualization) for further analysis, without building models for specific applications, which makes it difficult to achieve satisfactory results. For the specific application of network community discovery, a deep auto-encoder clustering model that combines community structure optimization for low-dimensional feature representation of nodes was proposed, namely Community-Aware Deep Network Embedding (CADNE). Firstly, based on the deep auto-encoder model, the node low-dimensional representation was learned by maintaining the topological characteristics of the local and global links of the network, and then the low-dimensional representation of the nodes was further optimized by using the network clustering structure. In this method, the low-dimensional representations of the nodes and the indicator vectors of the communities that the nodes belong to were learnt at the same time, so that the low-dimensional representation of the nodes can not only maintain the topological characteristics of the original network structure, but also maintain the clustering characteristics of the nodes. Comparing with the existing classical network embedding methods, the results show that CADNE achieves the best clustering results on Citeseer and Cora datasets, and improves the accuracy by up to 0.525 on 20NewsGroup. In classification task, CADNE performs the best on Blogcatalog and Citeseer datasets and the performance on Blogcatalog is improved by up to 0.512 with 20% training samples. In the visualization comparison, CADNE molel can get a low-dimensional representation of nodes with clearer class boundary, which verifies that the proposed method has better low-dimensional representation ability of nodes.

To achieve medium level visual region texture substitution in multi-regions contour curves, a multi-region image reconstruction algorithm was proposed. Firstly, the segmented region and index list of original image were extracted. Secondly, the polynomial coefficients representing region structure feature were got by piecewise iterative curve fitting and the smallest region texture sample was selected. Finally, the reconstructed region contour and synthesized region texture were used to achieve regional restoration. The proposed algorithm can be done successfully for images with massive textures. The experimental results show that it can reduce the amount of data required by image reconstruction and be applied to reconstruct the still background of image and video.

In order to eliminate the influence caused by fluctuation of Received Signal Strength Indicator (RSSI) and unreliability of individual beacon nodes in complex indoor environment, an indoor localization algorithm based on threshold classification and signal strength weighting was proposed. First, the reference points were classified and corresponding thresholds were determined according to the characteristics of indoor pathloss, then the received signal strength was used as reference weight to locate. The experimental results show that this algorithm can effectively reduce the error caused by RSSI random jitter, weaken the influence of individual beacon nodes which are disturbed, and improve localization accuracy.