Power system load forecasting model plays an important role in all aspects of power system planning, operation and control. Therefore, accurate power load forecasting provides an important guarantee for the stable operation of the power grid system. This paper first analyzes the current status of power system load forecasting, and finds that there are still some deficiencies in the existing forecasting models. In order to make up for these shortcomings, this paper proposes construction of short-term and mid-term power system load forecasting models based on hybrid deep learning. In the data preprocessing part, this paper proposes to use the exponential weight moving average method to process missing values. The detection method of abnormal value is the GeneralizedESDTestAD(GESD). This paper analyzes the historical load data of a regional power grid and four industries, and proposes a short-term power system load forecasting model based on Bi-directional Long Short-Term Memory(BiLSTM); For mid-term load forecasting, this paper first uses random forest and Pearson correlation coefficient to select features. Then a hybrid deep learning model is constructed based on BiLSTM and random forest. After optimizing the parameters of the model, a mid-term power system load forecasting model based on hybrid deep learning is constructed. Finally, the benchmark models are selected for comparative experiments. The experimental results show that the MAPE of the proposed model is 2.36%, which is better than the benchmark models. This proves that the proposed hybrid model can effectively improve the accuracy of power system load forecasting.

In order to ensure the security of computer systems and networks, it is very important to design and implement intrusion detection systems that can detect and mitigate network attacks and threats. Deep learning has great advantages in processing complex, high-dimensional and large-scale traffic data. Therefore, intrusion detection system based on deep learning method has better detection effect. Through the analysis of the research status, this paper finds that there are some problems in the existing intrusion detection system. To solve the problems of low detection accuracy, structure to be optimized and high false positive rate, this paper presents a hierarchical intrusion detection model which combines multiple deep learning models with attention mechanism. The advantages of this hierarchical model include: Firstly, the SCDAE model is used to extract the features of traffic data and reduce noise; Secondly, CNN is used to extract spatial features of network traffic data from the spatial dimension; Thirdly, BiLSTM is able to fully consider the relationship between the front and back features, so that the temporal features of network traffic data can be mined; Fourthly, a Self-Attention mechanism is added to weight the output of each time step to sum up and retain the important information in it. Thus, a CNN-BiLSTM-Attention model is constructed; Finally, the Softmax classifier is used to obtain the classification results. To verify the effectiveness of the proposed model, four time-sensitive and representative datasets are selected for experiments and five classical detection models are compared in this paper. The experimental results show that the classification accuracy of the proposed model reaches 93.26 % and the false positive rate reaches 7.53%.

This paper firstly describes the research status of online review text mining and finds out the problems existing in the mining and application of tourism texts. Aiming at these problems, this paper proposes a text mining method for tourism online reviews based on natural language processing and text classification technology. The first step is to analyze the validity of the online review text; the purpose is to remove the invalid text and improve the mining efficiency of the online review text. The second step is to conduct a comprehensive evaluation of scenic spots and hotels based on text classification technology and sentiment analysis. The comprehensive evaluation indicators are established for the five core service contents. High‐quality scenic spots and hotels are selected according to the ranking of comprehensive evaluation. The third step is to propose a mining method of tourism hot words based on natural language processing for the selected high‐quality tourist locations. The obtained hot words can intuitively show the impression of tourists on the scenic spot. The fourth step is to use mutual information combined with the left and right entropy to discover new words and to mine service characteristics of high‐quality scenic spots and hotel from the new words. Finally, the proposed new methods are tested on the crawled tourism online review texts. The experimental results show that the novel comprehensive evaluation method proposed in this paper can truly and objectively select high‐quality scenic spots and hotels and provide an important basis for the decision‐making of tourism management. On this basis, hot words and new words can be effectively excavated from relevant online review texts, and travel impressions can be fed back from various aspects and angles.

As an important infrastructure in the era of big data, the knowledge graph can integrate and manage data resources. Therefore, the construction of tourism knowledge graphs with wide coverage and of high quality in terms of information from the perspective of tourists’ needs is an effective solution to the problem of information clutter in the tourism field. This paper first analyzes the current state of domestic and international research on constructing tourism knowledge graphs and highlights the problems associated with constructing knowledge graphs, which are that they are time-consuming, laborious and have a single function. In order to make up for these shortcomings, this paper proposes a set of systematic methods to build a tourism knowledge graph. This method integrates the BiLSTM and BERT models and combines these with the attention mechanism. The steps of this methods are as follows: First, data preprocessing is carried out by word segmentation and removing stop words; second, after extracting the features and vectorization of the words, the cosine similarity method is used to classify the tourism text, with the text classification based on naive Bayes being compared through experiments; third, the popular tourism words are obtained through the popularity analysis model. This paper proposes two models to obtain popular words: One is a multi-dimensional tourism product popularity analysis model based on principal component analysis; the other is a popularity analysis model based on emotion analysis; fourth, this paper uses the BiLSTM-CRF model to identify entities and the cosine similarity method to predict the relationship between entities so as to extract high-quality tourism knowledge triplets. In order to improve the effect of entity recognition, this paper proposes entity recognition based on the BiLSTM-LPT and BiLSTM-Hanlp models. The experimental results show that the model can effectively improve the efficiency of entity recognition; finally, a high-quality tourism knowledge was imported into the Neo4j graphic database to build a tourism knowledge graph.

In recent years, the financial fraud event of listed companies has occurred continuously. The financial fraud has brought huge losses to the capital market and investors, hindering the investment allocation mechanism of the capital market. The current financial fraud prediction model can judge the company that may conduct financial fraud in advance. So, this can reduce economic losses. The key factor to construct the financial fraud prediction model is how to select the evaluation indicators. This paper analyzes the existing indicators selection method and finds the problem of low prediction accuracy. A key indicators selection method of prediction model based on machine learning hybrid mode is proposed. First, the contribution degree of the selected algorithm and model is ranked according to the features. The support vector machine with good classification effect and heterogeneity with other models is used as the intermediate evaluation model. A variety of selected indicators from machine learning are tested for AUC on the intermediate model. Well-performing machine learning models are selected and combined into multiple hybrid modes. These hybrid models are tested for AUC again. Experiments demonstrate that the hybrid mode of Lasso method and random forest performed best in the AUC test. The repetition indicators of the hybrid model are then selected as important indicators of the prediction model. Finally, the correlation of the indicators is tested, and the indicators beyond the threshold are removed. The selected key indicators effectively improve the accuracy of financial fraud prediction.

Financial data fraud by listed companies has brought an extremely bad impact on the market and society. Predicting the financial data fraud of listed companies in advance may reduce losses. Therefore, the key to solving the problem is to build a financial fraud prediction model. This paper analyzes the prediction and identification models of financial fraud at home and abroad in detail, and finds the problems existing in these prediction models. In view of these shortcomings, this paper proposes to build a financial fraud prediction model based on a machine learning fusion model. The first is the unbalanced processing of data samples. The oversampling method is used to improve the model prediction effect by setting a reasonable sampling ratio. Then, four machine learning models (GBDT, random forest, support vector machine, and decision tree) are selected suitable for financial data. The training set is used to optimize the hyperparameters of the four machine learning models separately. This paper proposes integrating the random search and grid search mechanisms to adjust the parameters to the optimum. Finally, a financial fraud prediction model is constructed based on the multimodel fusion of the integrated learning framework. First, the base learner integrates the predicted results of the four models and performs five-fold crossvalidation on the training set. The meta-learner then uses the GBDT model to train integrated data from the first layer, resulting in a fusion model. The experimental results show that the AUC value of the fusion model is significantly higher than that of the single model. Therefore, the fusion model proposed in this paper can effectively improve the prediction effect.