Search

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.

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.

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.

Safety production refers to the accident prevention and control measures in the production and operation activities, in order to avoid causing personnel injury and property damage accident. This paper firstly analyzes the performance and function of safety production supervision and management system, and points out the problems existing in the existing security supervision and management system. Big data analysis technology is an effective way to solve these problems and can be timely and accurate detection of hidden dangers, improve the ability of the investigation and management. This paper presents design of safety production supervision and data management system based on cloud platform. The system is the unified support system platform based on cloud computing and intelligent terminal technology, and can improve the efficiency of data collection and realize the prediction and early warning of the real by big data technology.

Based on the analysis of a double-layer vibration suppression bilinear system, the limitation of the classic passive vibration suppression technology was solved in case of relatively complex input force mixed signals. Using the maximum value principle and the constrained gradient method, the optimal damping semi-active control (ODSAC) curve of this system model was obtained. Due to the strong nonlinear characteristics of the controlling curve and even a non-derivable problem being existed, a pulse function was introduced, and a theoretic deduction was performed. The defects, that a curve with non-derivable points could not be solved, were overcome so that the proposed method was applicable to a bilinear system model. The stability of the controller was discussed in detail. Finally, the effects of the vibration suppression of the ODSAC method were compared with that of the passive damping system and that of the semi-active skyhook damping system under the excitation of the mix signal of single-frequency sine and pulse input signal, the mix signal of random and sine signal, the mix signal of random and impact input signal. The experiments of electro-rheological fluid (ERF) damper with single damping duct and simulation results show that the ODSAC strategy of a double-layer vibration suppression bilinear system is the best of the five kinds of vibration suppression effect of control strategies, and the vibration reduction effect with respect to the random and shock input mixed signal is remarkable, the vibration suppression effect of the new method is satisfactory.

The search engine is mainly to the user request information carries on the automatic information collection. Web crawler is an important part of search engine, it is an automatic extraction of web pages for search engines to download web pages from the web. The search system faces hundreds of millions of pages across the Internet, and there are several search servers in each data center, and there may be several crawlers deployed on each grab server. This constitutes a multithread distributed web crawler search system. The paper presents construction of search engine system based on multithread distributed web crawler.

Big data computing model refers to various high-level abstract models extracted and established from the diverse big data computing problems and requirements according to big data‘s different data characteristics and computational characteristics. In order to improve the data quality, a method of adding semantics to the data is proposed to help users identify big data’s pattern. In this way, we can make use of the semantic value of the data, detect big data, and propose a big data computing model through the semantic analysis of the data. The paper presents research on big data computing model based on semantic data analysis technology.

Crawler technology is a technology that uses web crawlers to capture information on a web page to get valuable information. The web crawler is a program that automatically extracts web pages. It downloads Web pages from WEB for search engines. It is an important part of the search engine. This paper introduces the architecture of distributed Hadoop platform, and by the principle of the web crawler puts forward design scheme of distributed crawler based on Hadoop and P2P technology, and ultimately determine the system layout of distributed crawler, module partition and distributed crawler flow control and power can be realized.

Clustering and classification are two of the most important technology in data mining; they were respectively used to detect data and decision space to feature space dependent. The paper describe analysis of intelligent association rule extraction technology based on big data mining method firstly, then discusses application of cloud computing and Internet of things technology in intelligent security. The paper presents application of big data mining technology in intelligent safe production on cloud computing platform. The final experiment shows that the proposed method of big data mining is very effective.

With the development of semantic Web technology, it provides a better solution to solve this problem, as the core of the semantic Web; ontology is a conceptual model of a clear specification. It provides a strong support for semantic similarity computation. This paper describes the semantic annotation method in the semantic web. This paper analyzes the construction of conceptual ontology in the semantic web. The paper presents improvement of semantic annotation method in semantic Internet of things based on conceptual ontology model.

The topic crawler filters the topic independent links according to a certain web page analysis algorithm, keeps the topic related links and puts them into the URL queue to be fetched. Then, according to a certain search strategy, the next page URL is selected from the queue and repeated until a certain condition of the system is reached. The best priority search strategy predicts the similarity between candidate URL and target pages according to a certain web page analysis algorithm, and selects one or more URL which is the best evaluation to grab. The crawler algorithm based on Web page extension is to evaluate the web page or website with indirect relation through known web pages or data. The paper presents research on topic crawler strategy based on web page extension and best priority search algorithm.

Big data technology refers to the rapid acquisition of valuable information from various types of large amounts of data. It can be divided into 8 technologies: data acquisition, data access, infrastructure, data processing, statistical analysis, data mining, model prediction and results presentation. The paper presents improved statistical analysis method based on big data technology. A statistical analysis model in big data environment is designed to extract useful information features from large amounts of data based on the Hadoop system by using its distributed storage and parallel processing mechanism.

The amount of information on the Internet is getting larger and larger, and the energy of the consumer and the ability to deal with information is limited. Electricity supplier enterprises in the development process to do are to use big data for personalized shopping guide. This paper analyzes the situation of the development of e-commerce industry in the background of big data, and puts forward the improvement method. The paper presents novel model of e-commerce marketing based on big data analysis and processing.

Particle Swarm Optimization algorithm is based on iterative optimization tools, system initialization of a group of random solutions, through iterative search for the optimal value. The basic idea of the fuzzy C-means clustering algorithm is to determine each sample data belonging to a certain degree of clustering, and the degree of membership of sample data is grouped into a cluster. Favor optimal solution in the sense of multi-objective particle swarm algorithm is efficient search capabilities. The paper presents the research of Building Fuzzy C-Means Clustering Model Based on Particle Swarm Optimization. Fuzzy c-means clustering is determined membership to each data point belongs to a cluster of a clustering algorithm. Particle Swarm Optimization is the process of the simulated social animals foraging moving group activities work of individual and group coordination and cooperation. DOI: http://dx.doi.org/10.11591/telkomnika.v11i12.3680