Research on Link Prediction Algorithms Based on Multichannel Structure Modelling

Today’s link prediction methods are based on the network structure using a single-channel approach for prediction,and there is a lack of link prediction algorithms constructed from a multichannel approach,which makes the features monotonous and noncomplementary.To address this problem,this paper proposes a link prediction algorithm based on multichannel structure modelling(MCLP).First,the network is sampled three times to construct its three subgraph structures.Second,the node representation vectors of the network are learned separately for each subgraph on a single channel.Then,the three node representation vectors are combined,and the similarity matrix is calculated for the combined vectors.Finally,the performance of the MCLP algorithm is evaluated by calculating the AUC using the similarity matrix and conducting multiple experiments on three citation network datasets.The experimental results show that the proposed link prediction algorithm has an AUC of 98.92%,which is better than the performance of the 24 link prediction comparison algorithms used in this paper.The experimental results sufficiently prove that the MCLP algorithm can effectively extract the relationships between network nodes,and confirm its effectiveness and feasibility.

Qinghai Embroidery Classification System and Intelligent Classification Research

.Qinghai embroidery is an artistic treasure of folk embroidery in Qing-hai Province.Classifying them to understand the differences between them is an important task.However,currently,there is a lack of a systematic classifi-cation method for Qinghai embroidery.First,by studying the history of Qinghai embroidery and a large number of Qinghai embroidery patterns,this paper divides Qinghai embroidery patterns into three categories:animals and plants,auspicious meanings,and geometric decoration.This method breaks through the regional classification system for the Qinghai embroidery.Second,this article utilizes four CNN models to classify the Qinghai embroidery datasets,exploring the differ-ences in the classification of Qinghai embroidery by different models andfinding the optimal classification model.The results show that the GoogLeNet model per-forms the best in the classification of Qinghai embroidery images,achieving the highest accuracy rate.This is mainly due to the small size of the Qinghai embroi-dery datasets and the application of the Inception structure and batch normalization technology in the GoogLeNet model,enabling it to better extract features and clas-sify Qinghai embroidery images.Through this research,we can provide a certain reference and assistance for the classification of Qinghai embroidery images and provide technical support for the protection and inheritance of cultural heritage.

Text-enhanced network representation learning

Network representation learning called NRL for short aims at embedding various networks into low-dimensional continuous distributed vector spaces.Most existing representation learning methods focus on learning representations purely based on the network topology.i.e.,the linkage relationships between network nodes,but the nodes in lots of networks may contain rich text features,which are beneficial to network analysis tasks,such as node classification,link prediction and so on.In this paper,we propose a novel network representation learning model,which is named as Text-Enhanced Network Representation Learning called TENR for short,by introducing text features of the nodesto learn more discriminative network representations,which come from joint learning of both the network topology and text features,and include common influencing factors of both parties.In the experiments,we evaluate our proposed method and other baseline methods on the task of node classihication.The experimental results demonstrate that our method outperforms other baseline methods on three real-world datasets.

Network Representation Based on the Joint Learning of Three Feature Views

Network representation learning plays an important role in the field of network data mining. By embedding network structures and other features into the representation vector space of low dimensions, network representation learning algorithms can provide high-quality feature input for subsequent tasks, such as network link prediction, network vertex classification, and network visualization. The existing network representation learning algorithms can be trained based on the structural features, vertex texts, vertex tags, community information, etc.However, there exists a lack of algorithm of using the future evolution results of the networks to guide the network representation learning. Therefore, this paper aims at modeling the future network evolution results of the networks based on the link prediction algorithm, introducing the future link probabilities between vertices without edges into the network representation learning tasks. In order to make the network representation vectors contain more feature factors, the text features of the vertices are also embedded into the network representation vectors. Based on the above two optimization approaches, we propose a novel network representation learning algorithm, Network Representation learning algorithm based on the joint optimization of Three Features(TFNR). Based on Inductive Matrix Completion(IMC), TFNR algorithm introduces the future probabilities between vertices without edges and text features into the procedure of modeling network structures, which can avoid the problem of the network structure sparse. Experimental results show that the proposed TFNR algorithm performs well in network vertex classification and visualization tasks on three real citation network datasets.