In solving the problem of feature weight calculation for automatic text classification, we use the most widely used TF-IDF algorithm. Although the algorithm is widely used, there is a problem that the feature categories have different weights when calculating the weights. This paper proposes an improved TF-IDF algorithm (TF-IDCRF) that takes into account the relationships between classes to complete the classification of texts. By modifying the calculation formulas of IDF to correct the problem of insufficient classification of feature categories, the naive Bayes classification algorithm is used to complete the classification. Finally, the proposed algorithm is compared with two others improved TFIDF algorithms. The results of the three text classification evaluation indicators show that the proposed algorithm has certain advantages in text classification.

This paper aims to overcome the defects of the existing multi-label classification methods, such as the insufficient use of label correlation and class information. For this purpose, an improved probabilistic neural network for multi-label classification (ML-IPNN) was developed through the following steps. Firstly, the traditional PNN was structurally improved to fit in with multi-label data. Then secondly, a weight matrix was introduced to represent the label correlation and synthetize the information between classes, and the ML-IPNN was trained with the backpropagation mechanism. Finally, the classification results of the ML-IPNN on three common datasets were compared with those of the seven most popular multi-label classification algorithms. The results show that the ML-IPNN outperformed all contrastive algorithms. The research findings brought new light on multi-label classification and the application of artificial neural networks (ANNs).