Applied Mechanics and Materials Vols. 488-489

Abstract: With the rapid growth of social media, recommendation for social activities is urgently needed to overcome information overload. Micro-blog, as one of most popular social media platform, has not a good enough recommender approach to satisfy users expectation. In this paper, we proposed a social recommender system using both exponential random graph model and sentiment similarity. Firstly, we built a good fitted graph model that was used to predict the probabilities of non-linked nodes. Moreover, we collected contents of each user for mining their emotions and select 106 features. Karhunen-Loéve transform (KLT) was chose to analyze the features of those texts. Based on KLT, average distances of text features were used to calculate the sentiment similarity. Therefore, according to the resort of similarities, we gave top-N recommendation for user. Finally, we studied this social recommender system on diabetes micro-blog. The metrics showed that our proposed social recommender system outperform other methods.

Abstract: This paper which starts from the characteristics of human social trust relationship, researches the trust in cloud computing environment. At present, the commercial cloud environment exist two main safe problems: one is the cloud platform trust security problems in their own special environment, the second is the security issues on the network of cheating, false, malicious ads and spam trust security . Aiming at solving this problem, on the basis of the existing trust models, this paper studies the cloud environment trust models according to the characteristics of the cloud environment, constructs a model of trust in the cloud environmentReputation and Risk Evaluation-Trust Model (RRE-Trust), and improves the trust evaluation algorithm. Finally, the experiment results show that, compared with the existing trust models, the proposed evaluate method of this model is closer to the real service behavior of service providers, resist effectively all kinds of malicious behaviors, shows a good robustness and ability in recognize.

Abstract: Extracting information from human resource information by using data mining technology in order that preserve and manage. Through java programming technique, the function of Arachnid that traversing the web page and extract web page content can be realized. Segmentation of content based on the technology of Chinese word segmentation machine of Chinese academy of sciences. Extract human resource information based on key word and save it in MySQL database, using the language of python to program the system of human resource information. The system can provide evidence for government decision-making.

Abstract: Complex ability evaluation of university students is an important means of training talents to the related teaching department, and it has a very important role to improve the training quality of university and the quality of teaching. In this paper, by analyzing of the disadvantages of traditional evaluation method, an evaluation model is put forward a kind of based on fuzzy comprehensive evaluation of university student' complex ability, and give the evaluation model of the specific composition and the corresponding algorithm. The model is better than common evaluation tools to weigh the various factors in the case of making information not lossing. It can naturally deal with the initiative and fuzziness of the human mind, so it makes the every student evaluated get an objective and fair evaluation result.

Abstract: This paper focuses on the architecture, functional structure and database design of JSP-based teaching management system discusses the implementation of course notice module, and achieves full functions of teaching management system by means of website construction.

Abstract: With the improvement of the power users requirements for power quality, traditional electricity pricing model cannot meet the needs of the users. At first, a concept of power user satisfaction was proposed, and the mathematical model of different categories of power user satisfaction was established. On the basis of comprehensive analysis of various kinds of power load characteristics, the mathematical model of the relationship between power and electricity price was given, and the simulation analysis shows that in the case of certain electricity price, power supply returns will be increased considering user satisfaction. Finally, an electricity price design mathematical model considering both user satisfaction and power supply enterprise's returns was established, and the model will provide certain theoretical basis to formulate a reasonable electricity price.

Abstract: In order to study the traditional accident causation theories and models about human factors in flight accident. A new aviation accident causation model based on the improved HFACS is constructed. Through using system aging, quality and Structure Entropy formula the degree of order of model 1 and model 2 are evaluated, and by selecting aging entropy, maximum timeliness entropy, timeliness, quality entropy and maximum quality entropy the structure of the degree of order is calculated to verify structure in the right to re-bias stability under the conditions. It is shown that the model 2's structure entropy are lower, aging and model quality are also larger, so the improved model is better structural model.

Abstract: As e-business develops rapidly, more and more product information and product reviews are posted on the Internet. These contents will have a great significance for companies and consumers. This paper focus on customer reviews of product, and construct a technology oriented research framework for the sentiment analysis. Further more an improved theoretical framework of aspects extraction is proposed, which based on products feature mining issues from customer reviews. This two theoretical framework can help researchers acquire supported valuable data for additional researches including the study of behavioral.

Abstract: The purpose of Decision Problem Analysis is to identify methods of solving decision problems, which can make it facilitate for the user to make decisions and solve problems quickly in Decision Support System. In order to improve the intelligence and accuracy of Decision Problem Analysis, an Ontology-based method is proposed. The method designs the ontic conceptual structure of decision problems by using Ontology. Finally, an instance was presented to explain how to use the method, and to verify the effectiveness of the method.

Abstract: Considering the feature of round steel enterprises with the contradiction between mass production size and the multi-variety and small-batch market demands, an order-grouping problem is studied with transportation optimization. Two objectives that to minimize the delivery destination difference and delivery date difference are presented, and an algorithm improved k-means algorithm procedure is proposed to solve the problem. In this algorithm, a method is applied to generate representative initial centers of k-means algorithm for optimization. Experiment results show that the algorithm is feasible and effective.