Earticle

In the social tagging systems, tags are results of use annotations. Combining social tagging with personalized information recommendation, this paper designs a personalized information recommendation method based on social tagging on the basis of the existing method of personalized information recommendation. This method clusters user tags and resources tags to form tagging clouds, and then calculate the similarity between the user community tagging cloud and resources community tagging cloud to make personalized recommendation. The experiment results show that the method has achieved a good expected effect and can raise the accuracy and efficiency of personalized information recommendation.

The Internet is one of the most widely available services in the world today. With the Internet, people are now looking for reviews on the Internet; more specifically, the social networking services. Within the social network medium, we can identify a suitable service that describes more about a person’s personality as the subject. The growth of social networking popularity has contributed to the in-crease in information available on social networking services. The flexibility of these services allows writing individual thoughts without restrictions. With the vast information available on social networking sites today, how is it possible to look all of these opinions? How do we know which opinion holds truth? How do we know if someone is not bias based on his writing? Hence, it is seen necessary to filter opinions. In this paper we look at the possibility of using search ranking as a medium of filter opinions by exploring opinion mining methods, social net-working candidates and search ranking methods. With existing sentiment analysis techniques, we can obtain opinions that are then ranked against a set of key-words.

For the Chinese ancient art, stone carving experiences the etching of the natural environment and become blurred. How to process and store their images becomes a quite important problem. In this paper, an image process and storage method have been proposed. In this method, the blurred image will be firstly pretreated. Then, relative images which have high similarity with the blurred image will be retrieved in the database. Thirdly, blurred images are processed with template matching method to improve the clarity. Finally, processed images and the original images are stored in the dataset with classification. A blurred art image is used as the example to verify the validity of the method. The blurred image contains a person and a door. The query method is proposed with higher speed and precision, which is due to the pretreatment and mixed subgraph method. The pretreatment can improve the precision of the query content, while the mixed subgraph method can provide the completeness of the query content. Both the mechanism improves the retrieval precision. The blurred images excavated in the soil are often with pitting and rust, which is corresponding to the noise in the image process. In the image process step, the image is fixed up when the blurry points are big enough. The verification confirms the validity of the method.

Representing words as continuous vectors enables the quantification of semantic relationships of words by vector operations, thereby has attracted much attention recently. This paper proposes an approach to combine continuous word representation and topic modeling, by encoding words based on their topic distributions in the hierarchical softmax, so as to introduce the prior semantic relevance information into the neural networks. The word vectors generated by our model are evaluated with respect to word relevance and the document relevance. Experimental results show that our approach is promising for further improving the quality of word vectors.

Normal human Beings can communicate with each other with the help of various different languages, however, the people who can’t speak have different sign languages to communicate with others. But a major setback of sign language is that only people those who know sign language can communicate with them. Individuals habitually use gestures to interconnect. Gestures are used for directing to an individual, to get his devotion and convey statistics about temporal and spatial physiognomies. Gesticulating does not only elaborate verbal language, but it is part of the language production practice. In our paper we recommend a scheme which will be able to convert the sign language into words or sentences. The system takes inputs several hand gestures then processes its meaning and verifies it with the stored gestures in database and results the corresponding output on the screen.

The hidden topic model of Chinese text, which possesses complicated semantics, is urgently needed, since China has occupied an increasingly significant role during the booming development of globalization over recent years. This paper details and elaborates the basic process of extracting latent Chinese topics by demonstrating a Chinese topic extraction schema based on Latent Dirichlet Allocation (LDA) model. Furthermore, the application was practiced in CCL, an authoritative Chinese corpus, to extract topics for its nine categories. With rigorous empirical analysis, extracting the LDA results has a considerably higher average precision rate as opposed to other three comparable Chinese topic extraction techniques; however the average recall rate is worse than KNN and almost the same with the PLSI model. Moreover, the recall rate and precision rate of LDA-CH is worse than LDA-EH. Therefore, the LDA model should be improved to adapt to the distinctive feature of Chinese words with the purpose of making it better for Chinese topic extraction.

This study uses the system dynamics tool Vensim to conduct simulation on two different types of maintenance spare parts information sharing models (i.e. the inter-user joint maintenance spare parts management model and the supplier-involved joint maintenance spare parts management model) as well as the spare parts management model without the implementation of information sharing. By comparing the change of revenues of both the supply and the demand sides before and after the implementation of information sharing, as well as under different information sharing models, we propose the optimal way of information sharing, and also contribute managerial implications for the implementation of information sharing in maintenance spare parts inventory management.

In the era of big data, the value of Internet public opinion hot spot extraction is particularly prominent. In order to further develop the application value of Internet public opinion hot extraction, this paper constructs the Internet public opinion hot spot extraction framework, and puts forward an Internet public opinion hot spot extraction method VM-Rep based on voting mechanism and MapReduce framework. A heuristic algorithm based on voting mechanism is proposed for the hot spot of Internet public opinion, and MapReduce is used to improve the ability and efficiency of processing massive data. Experimental results show: VM-Rep’s coverage is significantly better than Top-k, K-means and Agglo, and the redundancy of VM-Rep is the least; VM-Rep takes the least time in the four methods, embodies the advantages of VM-Rep method for massive data.

Based on economic growth theory of modern economic theory, with Fujian Province as object of study, panel data from 1981 to 2013 is selected in this paper. Besides, VAR model is established. Co-integration test, Granger causality and impulse response function are used for quantitative analysis on correlation between consumption, investment, import, export and economic growth in Fujian Province. With Eviews8 software as analytical tool, the empirical study conducts analysis from the perspective of econometrics and tests the stationarity of unit root data; determines optimal lag intervals through VAR model analysis on different lag intervals; determines the existence of co-integration relationship of data through Johansen co-integration test; tests the causality of data with Granger causality test; conducts stationarity test by establishing VAR model; studies the variation trend of influence of consumption, investment and import and export on economic growth in the next ten years through impulse response. The author reveals long-term stable balance among consumption, investment, import, export and economic growth in Fujian Province as well as bidirectional causal relationship between consumption and economic growth. For a short term, household consumption is of relatively strong positive impact effect on economic gain, which is served as major impetus of economic growth. Domestic investment is of relatively obvious promotion for economic growth in a short term and of certain negative effect on economic growth in a long term. Import and export have significant function for promoting economic growth and profound influence in Fujian as coastal city. On this basis, the author proposes some policies and recommendations for macroeconomic regulation and control in Fujian Province.

In this study, a work information classification system for the TBM tunnel construction was proposed along with a conceptual model of an automated cost estimation system, with the goal of relieving the cumbersome operation of existing systems. The conceptual modeling of the automated cost estimation system involves systematizing the construction cost estimation process, by organizing a work information classification system in the form of a relational data base which reflects the processes that comprise the construction work. This proposed automation system is expected to be effectively applied to TBM tunnel construction sites as a scientific cost estimation system.

We propose JAS, a JVM-based active storage framework for object-based storage systems. JAS programs the active storage functions of users as Java codes, and allows them to be executed on different OSD platforms (Operating systems and hardware) without recompiling. JAS offloads the active storage code from a client to the OSD by extending the standard OSD command set, and execute the Java code on the OSD on-demand by triggering them through extended client interfaces. We have implemented JAS under an object-based storage system. Experimental results show that the JVM-based active storage framework has been successfully set up, and this cross-platform design can largely improve system performance.

Privacy preserving data mining (PPDM) has become a popular research direction in data mining. Privacy preserving data mining is an approach to develop algorithms by which we can modify the utility values of original data using some techniques in order to protect sensitive information from unauthorized user. Protecting data against illegal access becomes a serious issue when this data is required to be shared onto the network due to some reasons. To hide the sensitive information, many approaches have been proposed. In this study, we are proposing an efficient method, for protecting high utility itemsets using distortion technique where the values for high utility items are altered to achieve the privacy. Algorithm is designed in such a way so as to handle privacy without disclosure of sensitive information. The algorithm can completely hide any given utility items by scanning data iteratively. The results when compared with existing one show significant reduction in execution time.

Data analysis on environmental factors is crucial to aquaculture. Several significant parameters (temperature, pH and dissolved oxygen) and factors related to it are discussed in this paper. Data preparation including fixing some missing data and inaccurate records by non-linear function approaching in the first sampling process. Then the utilization of deterministic tracking theory is adopted for dynamic system analysis. Furthermore, time series analysis from different water layers based on this theory suits the real environment well, Radial Basis Function Neural Networks is well applied in tracking the parameters trend both globally and locally. The results provide effective references for systemic data analysis and control engineering.

The integration of informatization and industrialization is depth integration of informatization and industrialization in many fields, which is the new developing way for informatization and industrialization. It is a complicated process of integration for enterprises, which is necessary to learn from the experience of other enterprises to improve efficiency. But facing numerous cases, it is hard for enterprises to pick up the right cases, because of shorting of effective recommended algorithm in academic field and practical application. Base on the studying the existing recommend algorithm, this paper try to design new algorithm with analyzing the structure and characters of the cases. First of all, the case structure is analyzed and are classified and the information of cases are classified and graded according to the degree of integrity; secondly, AHP is used to give weights to different attributes; then, the recommended algorithm is illustrated; finally, the cases similarity are calculated and the high similarity cases are outputted to the users.

Frequent Itemsets Mining(FIM) is a typical data mining task and has gained much attention. Due to the consideration of individual privacy, various studies have been focusing on privacy-preserving FIM problems. Differential privacy has emerged as a promising scheme for protecting individual privacy in data mining against adversaries with arbitrary background knowledge. In this paper, we present an approach to exploring frequent itemsets under rigorous differential privacy model, a recently introduced definition which provides rigorous privacy guarantees in the presence of arbitrary external information. The main idea of differentially privacy FIM is perturbing the support of item which can hide changes caused by absence of any single item. The key observation is that pruning the number of unpromising candidate items can effectively reduce noise added in differential privacy mechanism, which can bring about a better tradeoff between utility and privacy of the result. In order to effectively remove the unpromising items from each candidate set, we use a progressive sampling method to get a super set of frequent items, which is usually much smaller than the original item database. Then the sampled set will be used to shrink candidate set. Extensive experiments on real data sets illustrate that our algorithm can greatly reduce the noise scale injected and output frequent itemsets with high accuracy while satisfying differential privacy.

A document classifier is an essential tool for classifying the various types of documents being generated in the Big Data era. In recent years, the wide variety of information services available for use with smartphones and portable mobile devices (tablets) have provided a technique that efficiently classifies the quality of sorted data. A common type of document classification scheme is the naïve Bayes classifier. The Naïve Bayes scheme is based on performance classification, which varies widely depending on the method of extraction used in the document. In this paper, we propose a system model that offers feature extraction methods which combine frequency with associated words. This model is then applied to the Naïve Bayes classifier to precisely classify documents. This method is proposed as an alternative to using traditional classification techniques. In addition, experiments will be evaluated by the existing document classification techniques and the proposed techniques.

As the Internet is becoming a virtual platform for people’s sharing information and communicating daily, more and more people are connected together and the contact area between their personal privacy and the outside world turn to be unprecedented large. Therefore, the users of the Internet are faced with an enormous privacy threat. Anonymous P2P technology is a good way to solve the above problems. The ultimate goal of all anonymous P2P systems is to obtain the anonymity. Anonymity assessment on different anonymous P2P systems can be used to compare the amount of anonymity offered by different systems or to adjust the parameter setting of the same system according to different need, which is of great guiding significance to improve anonymous communication systems. In this paper, from the angle of relationship anonymity between the sender and the receiver, we presents the BIG-MIX anonymity method, taking the paths cross of different messages into consideration, to make anonymity assessment on anonymous P2P networks with the use of mapping. And with this method, this paper makes a quantitative analysis on the relationship anonymity of anonymous P2P networks under the mechanism of Threshold Mixes, Timed Mixes or Pool Mixes.

Concerning the condition that there is a glittering array of disadvantages such as frequent candidate collection of Apriori algorithm, this paper comes up with cost-sensitive filtering matrix Apriori algorithm based on weighting. What’s more, with the help of FP-tree algorithm, we can carry out cost-sensitive learning through relevant data of its constructed decision tree to set different weighting for data and confidence level.

The object of opening a public data is to make them freely used by public so as to improve quality of civil life and activate new industry and job creation. Opening a public data is a policy that has been importantly emphasized around the world and active policy for public open data is creating successful cases. Public open data needs proper quality to achieve public objects. However, quality management and a lack of standardization cause error data found and a lack of availability, and a lack of guideline about quality. This research figures out quality characteristics necessary for public open data through analysis and verifies establishment of model through survey of professionals. Quality characteristics of public open data verified by survey are publicity, availability, credibility, and suitability. It has its significance for suggestion of quality characteristics necessary for improvement of quality and availability of public open data.

With the development of semantic web and ontology application, there is a large number of ontology whose scale is large and the structure is complex in different fields. The existing mapping method and mapping system perform well when dealing with the mapping between a lightweight small ontology. However, when comes to the large-scale ontology, it is full of challenges to the methods and systems。To this end, this paper proposes a method of ontology compression based on conceptual cluster to compress. Firstly, it calculates the semantic similarity and semantic correlation of ontology concepts with the DICE coefficient method and the information entropy technology to get semantic relation. Secondly according to the semantic relations, it carries on the conceptual cluster in the concept space so that the concept of semantic relations closely together in groups. The concept of cluster in space is reduced, and the "noise concept" which is independent of the mapping is removed, and the purpose of the large-scale ontology compression is realized. Experimental results show that the method is so effective that it can compress the volume of large-scale ontology in the mapping problems.

With the development of network technology, the data capacity become more abundant. How to effectively manage the data, the retrieval more quickly, accurately, improve the data classification accuracy becomes crucial. The BP neural network algorithm with its learning speed, strong ability to adapt and is widely used in network in data mining. Exist but its convergence rate is not high and big error and other shortcomings, therefore, on the basis of traditional algorithm, an improved BP neural network algorithm is put forward. Low error, through experimental analysis, the improved algorithm convergence rate is better.

This work is the first of a series of works that discussed a model for solving the problem of non-modularity in electronic voting systems. It analyzed and described the system from a structured and layered perspective; with the system layered in order to achieve modularity. First, was the description of similar models and architectures that have been previously proposed by other authors on related subjects and how good their models have been. Our proposed system was structured into three layers that include Application, Security Service and Network Access layers. However, in this work, focus was more on describing the Application layer, which is further split into two sub-layers, namely the Application Hardware and Application Software sub - layers. A couple of components and modules that make up this layer were carefully outlined, reviewed and discussed.

The social media is seemingly becoming a big influence in the politics and a core strategy for political campaigns. Social media has become a coordinating tool for almost all of the political movements globally. The prospective of social media is mainly in their support of civil society and the public sphere. Social networking sites have gradually become integral part of people’s everyday life. Internet, through the use of the social media platforms, is now being used to convey messages to a diverse audience more directly. Also, the Internet provides a major technological stimulus to the modernisation and professionalisation of political campaigns. This paper, based on document analysis and insights of data published on the social media platforms such as Twitter and Facebook, gives a brief discussion on the advent of the social media networks and the politics. Examples of the events where the social media have influenced certain activities or actions in the political landscape across the globe are also briefly cited and discussed in this paper. Some of the events highlighted in this paper which have been influenced by the social media turned out to yield positive outcomes whilst some outcomes are not positive. This paper is concluded with a view that the social media continues to have an influence in the political landscape. Thus, reactions, feedback, conversations and debates are generated online as well as support and participation for offline events. Messages posted to personal networks are multiplied when shared, which allow new audiences to be reached through the social media. However, this obviously does not mean that every political movement that uses these tools will succeed, because some state(s) have not lost the powers to react through banning them or censoring them.

With the remarkable progressions in network communications and modern day technology, Mobile IP became very significant as people nowadays have mobile devices with them. If one network goes down, the other connected networks are affected. However, Mobile IP makes this problem practically non-existing. Wherever the user’s location is with his/her mobile device, a unique IP address of its own exists consistent with their device. Mobile IPv6 advanced features compared to MIPv4 are optimized to reduce packet loss. Herein, issues on security threats alleviates in the mobility of wireless networks. This paper focuses on MIPv6 mobility support by proposing enhanced methodologies in providing secure communication.