The Internet combines economic system forming the innovative cyber economic system. The principal characteristics of cyber economy are discussed and the reason of its rapid development analyzed. From the perspective of information, cyber economy is of faster information growth, and the core reason of the growth is algorithms. As the foundation, algorithms prompt the new form of cyber economy i.e. cyber intellectual economy. Analogously, blockchain is expected to be the reliable infrastructure of cyber intellectual economy and consequently to innovate it. Furthermore, the characteristics of blockchain are also analyzed, pointing out the future opportunities and challenges.

The background of the Industrie 4.0 is first summarize, the EU's policies on Internet-of-things briefly reviewed. The core of IoT is to break the barriers among the information islands and to share the information, and therefore the IoT is the results of the informationisation and can be concluded as "to let the thing TALK". The techniques such as RFIDs, wireless sensing networks, advanced wireless communications, internet, etc. are the basics for realizing the informationisation. Finally "7Cs" principle has been proposed for IoT.

The learners who participate in the MOOC are classified by using of AHC and K-means, based on learners' learning behavior on MOOCs platform. Then, the impact of different learning behaviors on the MOOC's learning outcomes are studied and analyzed, and the course completion rate and course grade of different types of learners are compared and analyzed in detail by Chi-square test and one-way ANOVA. Finally, suggestions about how to improve the MOOCs learning effect and the design of the structure and content of MOOCs curricula are given. It also provides the reference for the further development of MOOCs education in colleges and universities.

The Belt and Road Initiative, Intelligent manufacturing 2025, Industrial revolution 4.0 and so on are reshaping our industrial development and upgrading. Facing the new demand of engineering professionals, it is imperative to construct first-class network engineering disciplines, exploring the cultivation of professional talents of high quality, together with the new engineering operations, engineering education professional certification, as well as excellent engineer pilot plan, etc. The new ideas and measures for the development and reform of network engineering discipline are discussed in six aspects. What's to be done is to achieve a new leap in the integration of production and higher education, continuously delivering excellent network engineering professionals for the country.

The explosive growth of audio streams brings difficulties in storage and transmission; however, many methods could not give high compression ratio while keeping the quality. In order to solve this problem, the proposed method compresses amplitude spectrum of voice by constructing a dynamic sparse voice dictionary based on incremental learning. It calculates amplitude envelopes spectrums via Short-Time Fourier Transform (STFT) firstly, and then it uses a dictionary to fit each envelope by projecting high dimensional vectors to several 2D planes. In addition, it minimizes the number of dictionary items and therefore can store the parameters of linear interpolation instead of spectrums. Otherwise, if the fitting step above fails, it will store this window of spectrum directly. By using dictionary and parameters of linear interpolation, it can reconstruct the spectrum efficiently in decompressing process. The results of experiments show that comparing with other methods, the proposed method gives high compression ratio as well as better accuracy in decompressing, and adapt to live voice stream encoding with high sampling rate.

Sentiment analysis has attracted interest recently. In financial applications, it can be a reference for investors. However, existing approaches are either so specific as to cause data drift or too general to be precise. Therefore, a general Chinese text classifier for online reviews and news on stocks is optimized. A corpus with 20000 items is first collected. Then, each item is labeled by three persons as ground truth. After that, the THUCTC is optimized, thus optimizing a general Chinese text classifier in three aspects. First, by tokenization, the THUCTC is modified to a 2-gram with a stemming dictionary method and got better results. Second, the best kernel is selected for classifier. The Liblinear kernel is found to be better for people pressed for time. On the other hand, the Libsvm kernel is good at promoting accuracy. Third, a finance-oriented sentiment dictionary is set based on Chi-square and TF-IDF approach. It can be used by on-the-shelf general text classifiers. In this way, the result can be generalized without the loss of preciseness.

The research of intelligent speech recognition technology has been going on for a long time. However, due to the characteristics of variability, instantness, continuity and dynamic of the speech signal itself, the identification of the speech still has some difficulties when the machine is put in different environments, especially in the noisy environment. In order to improve the recognition accuracy of the noisy speech signal, a commonly used noise estimation algorithm was studied, which was based on the time-averaged algorithm of posterior signal noise ratio. And an improved algorithm of the smoothing factor was brought up on the basis of the previous algorithm. The voice activity detection algorithm and the above two algorithms were simulated under different input signal-noise ratios. The comparative analysis of the operation results shows that the improved algorithm can improve the output segment SNR by 2.1 dB compared with the voice activity detection algorithm, and it can also improve the output segment SNR by 0.5 dB compared with the original time recursive average algorithm. It is indicated that the improved algorithm can effectively improve the quality and intelligibility of the speech signal at low input SNR.

In multi-instance learning, classification performance may be limited due to the noisy data or a scarce amount of labeled data. To solve this problem, an algorithm based on multi-task multi-instance anti-noise learning is proposed. On the one hand, in view of the noisy data, the algorithm trains a classifier by assigning the instances in bags with different weights. And the weights of instances are updated by adopting an iterative optimization framework which decreases the influence of the noisy data. On the other hand, in view of insufficient labeled data, the classifier is extended to multi-task learning to train multiple learning tasks at the same time, so that the performance of each learning task can be improved by sharing the classification information among the tasks. Extensive experiments have showed that the proposed classification framework outperforms the existing classification methods.

With respect to the cost of advertising investment being limited, the processing of rationalizing and maximizing marketing investment via online keyword bidding is hard. In order to solve this problem, an optimization method is proposed, which is based on Role-Based Collaboration (RBC) and its E-CARGO model. According to history data, it models the problem by mapping keywords and their combinations to roles and groups (Group Role Combination) and using linear programming to obtain the best investment prediction. The proposed methods are verified by simulation experiments. The experimental results present the practicability of the proposed solutions. Using the proposed methods, decision makers need only to provide investment budget. The maximal profitability, the rate of return and the investment range are obtained.

Online portfolio selection is an important research problem in the field of quantitative investment. To avoid the interference with current investment decisions caused by the stock price data far from now in the intensely fluctuating stock market, online portfolio strategies based on moving window are designed. Using the recent stock price data, the recent performances of all constant rebalanced portfolios are computed and ranked. An online portfolio strategy based on moving window is designed by weighted averaging all constant rebalanced portfolios. Further using adaptive learning method to select the length of the moving window, the adaptive learning strategy is put forward. Empirical analyses are made on the proposed strategies using the real stock price data. The results show that they have better performance.

According to application demand of industrial big data for smart factory, the big data cloud platform was built to achieve multi-source heterogeneous data acquisition in whole networks and whole processes, provide multi-level analysis scheme, establish data mining model library, and support the intelligent bearing production, network collaborative manufacturing and intelligent services. With the platform, petroleum refineries enhanced their core competitiveness.

As an advanced integration management platform of building information, BIM has been widely used in the international construction market, and is promoting unprecedented reforms in the construction industry. However, there are still many risks involved in the application of BIM technology in construction projects, such as legal and policy risks, industry risks, organization management risks, technology risks, economy risks, etc. Based on previous studies, the risks are summarized, and the main influential factors of BIM application risks and the causal relationship between different factors are analyzed depending on DEMATEL method. On one hand, a conclusion is drawn that most risks of BIM application lie in the investment income uncertainty of the enterprise from the centrality aspect. On the other hand, it can be concluded that the main factors from the reason degree are the BIM standard, guide defect and lack of domestic BIM software. Finally, some suggestions are provided to reduce the risks of BIM application, such as perfecting the BIM standards and application guidelines, strengthening the BIM technology research and development, improving the BIM contract management and business process, and building BIM professional team, etc.

The decision making process of PPP projects always focuses on the profitability and meeting the interest of government sectors and private investors. The decision maker paid little attention to the requirement and preference of the public users, resulting in low social satisfaction and poor national economic effect. In order to reduce the risk of social stability which always comes from project's low social satisfaction, the feasibility and effectiveness of the public participating in the decision-making of PPP projects are discussed. Then a multi-objective decision making model based on Bayesian theory is established to help the decision maker adopting the method of adding the public factors in the planning stage of PPP project. The simulation process is introduced and the simulation results are analyzed through an assumed commercial project. The results show that the public and investor's satisfaction are both considered in the decision making process of project implement scheme. The decision making system based on Bayesian model offers a quantitative way to assist the decision maker in getting a reasonable result with good public participation.

The efficient numerical methods for solving the American option pricing model under jump-diffusion is studied. First of all, the high accuracy compact difference scheme is applied to discrete the option pricing model in the spatial direction, and discrete the temporal variable of the resulting ordinary differential equation to the linear complementarity problem (LCP). The approximation value of option price is obtained by solving the LCP. Finally, in order to overcome the nonsmoothness of payoff function, the singularity separating method is utilized for the American option pricing model to improve the accuracy of calculation. Numerical examples demonstrate the superiority of the algorithm.

With the development of e-commerce in China, the traditional e-business service mode can no longer meet people's shopping needs. Personalized recommendation for customers is a problem worthy of study. In this research, the improvement of Apriori algorithm is used to mine user interest information and the user correlation. Then, a user behavior model is set up, and can recommend the goods of interest, and improve the user's purchase experience. Experiments show that the improved Apriori algorithm improves the accuracy and speed of the recommendation system.

Aiming at the big amount of data in the big data environment and the dynamical change in time, a migration algorithm for data based on cloud memory is proposed. Firstly, the big data in ocean environment is represented; the load prediction algorithm based on the prediction of the gray model, where the algorithm can predict the load for the next time based on the historical load information. The data migration algorithm for migrating the load among the servers is proposed on the condition of the setting the smallest and biggest load threshold. In the environment of the Cloudsim, the simulated result shows this method can balance the load in the ocean big data, with the high load balance and load balance efficiency. Compared with the other methods, it has a better load balance ability.

A method to deal with dispatch of Microgrid is proposed based on the dynamic fuzzy chaotic particle swarm algorithm. Using dynamic objective function and fuzzy theory to solve the defects of the subjective weight, a microgrid scheduling model is found which aims for lower maintenance cost and environmental pollution. Multi-objective dispatch of microgrid belongs to the multivariable and strongly nonlinear optimization problem. Since the traditional particle swarm algorithm tends to trap in the local superior, in the particle initialization, a chaotic mapping combining the Chebyshev maps and the Logistic map is introduced, and in the process of particle update, the Logistic map is introduced which increases the ergodicity of particles and strengthens global optimization ability of algorithm. According to the value of inertia weight in particle swarm updating, the strategy of changing the inertia weight with the gradient of iteration number adopted. The simulation experiment of multi-objective dispatch of microgrid shows that the algorithm has higher convergence speed and better convergence effect.

In order to solve the problem of the low level of the nature, intuition and accuracy of current mixed reality system interaction, a new method based on baton-like interaction is proposed. The key to its implementation is to measure and track the baton's attitude in real time and accurately. Therefore, the focus is on a two-color vision-based visual tracking method to measure the attitude of the baton. The specific idea is to perform fast contour extraction on the color mark, and then patch the contour by the minimum area rectangle method to get the posture of the color mark projection in the plane. Finally, the principle of binocular vision is used to calculate the three-dimensional pose of the baton. In order to verify the validity of the proposed method, experiments are carried out to analyze the real-time and accuracy of the proposed algorithm. The results show that both the real-time and accuracy of the proposed method can meet the interaction requirements.

To achieve the unified collecting, processing, analyzing and decision of the production and operation data for the Southern Transport Group, the production operation and decision-making system based on big data was constructed. It can help the administrators rapidly analyze the production and operation status of each department. With the help of the powerful visual function of the management cockpit, the decision makers can analyze the multidimensional datasets of the production and operation. It can also raise the level of informatization. On this basis, analysis and implements for sorts of big data were carried out to improve the intelligent level of the production and operation for the transport group. It can be useful for data mining and analysis of related study.