Liu Jinxing team of Computer Institute published the latest research results in top international journals.

Recently, the research team of Professor Liu Jinxing from the School of Computer Science of Qufu Normal University has made important progress in the field of artificial intelligence. The related research results are "n cplp: a novel approach for predicting microscopic-associated diseases with network consistency projection and label propagation", "predicting mirna-disease associations through deep auto encoder with multiple kernel learning" and "a new graph auto encoder-based consensus-guided model fo" Scrna-seq cell type detection ",published in IEEE transactions on cybernetics (IEEE TCYB) and IEEE transactions on neural networks and learning systems (IEEE TN NLS).Qufu Normal University is the first signatory of the paper, and Professor Liu Jinxing is the correspondent of the paper.

IEEE TCYB期刊是人工智能领域最具影响力的国际学术刊物之一，SCI一区，影响因子19.118，在Automation & Control Systems类别排序1/65，Computer Science, Cybernetics类别排序1/24。IEEE TNNLS期刊由美国电气和电子工程师协会创办，是人工智能顶级期刊之一，SCI一区，影响因子14.255，在Computer Science, Hardware&Architecture类别排序1/54。

A large number of clinical studies have confirmed that there is a close relationship between microorganisms and diseases. Therefore, it is particularly important to infer the relationship between potential microorganisms and diseases. Professor Liu Jinxing’s research team proposed a new method based on network consistency projection and label propagation to predict the relationship between microorganisms and diseases. Methods The medical thesaurus and 16S rRNA gene sequence were used to calculate the semantic similarity of diseases and the functional similarity of microorganisms, and the network projection scores of microbial space and disease space were obtained by increasing the network consistency projection. Finally, the disease-related microorganisms were accurately predicted by the tag propagation method.

The association between miRNA and diseases is an important part of preventing, diagnosing and treating complex diseases. Professor Liu Jinxing’s research team proposed a deep learning method to predict miRNA- disease association by using a deep automatic encoder with multi-core learning. Firstly, the multi-core learning method was applied to construct miRNA similarity network and disease similarity network respectively, then the integrated miRNA feature representation and disease feature representation were input into the deep automatic encoder, and finally the new miRNA- disease association was predicted by the reconstruction error method.

In order to make effective use of single cell data and better explore the heterogeneity among cells, Professor Liu Jinxing’s research team proposed a consensus guidance model based on graph automatic encoder. The model generates feature matrix through feature learning, and carries out similarity learning based on distance fusion method. The model can accurately identify key features and effectively save the internal structure of data, which improves the accuracy of cell type identification.

Professor Liu Jinxing is the head of the artificial intelligence research team of Computer College of Qufu Normal University. Focusing on the national strategy of "Healthy China" and the big scientific plan of "Precision Medicine", the team actively connects the two industries of "new generation information technology" and "medical care and health care" of the top ten industries in Shandong Province, aiming at the theory, algorithm and software related to the intelligent calculation of health big data, especially the disease genetics and clinical big data. The innovation team has undertaken 10 national natural science funds, with accumulated horizontal and vertical funds of more than 10 million yuan, and published more than 150 papers in high-level journals and academic conferences at home and abroad, including more than 50 SCI TOP papers.