Earticle

In this paper, on the basis of traditional phonetics, by using the methods of experimental phonetics and voice pattern theory, Analyzed and summarized the vowel pattern of the monosyllables in Uyghur language. The statistical analysis is carried upon the vowel formant frequency values in monosyllables, and discussed by using Joos method in more details. For the first time, with the actual experimental data proves the accordance of tongue location features of Uyghur vowel with the traditional knowledge from hearsay. The research results of this paper will have a high reference value for the study and application development of both Uyghur language and the other languages are belongs Altaic language family.

As the starting age of English education becomes younger around the world where they teach English as a foreign language, the debate on early English education is an unsettling issue as not only academic research, but also educational policy. To resolve the unsettled issues, a new approach is used: the big data and its analytics. To explore the pros and cons of the early English education, the study uses the analysis of research abstracts collected from scholars.google.com and www.kci.go.kr. It also analyzes the data posted on the discussion sites such as agora, daum café and naver café, plus daily interactions of the early English education using SNS. The study uses opinion mining technique using tools such as Sisense and WEKA to lay out the data and analyze them as basic data analytics. The results show that pro early English education is commonly co-occur with critical period, lateralization, ultimate attainment, universal grammar, fossili-zation, inhibition, acquisition process, bilingualism and exposition. Essays against early English education are related to no critical period, no authentic input, not effective, no universal grammar, national identity loss, self-identity loss and L2 interference. Other extraneous factors are based on practical problems such as social pressure, outcome pressure, political pressure, test reform and statistics.

Sentiment dictionaries or lexicons are core elements for “bag-of-word” approaches of opinion mining or sentiment analysis. Rather than using general-purpose sentiment dictionaries, domain-specific sentiment lexicons can contribute to improve performance because they can reflect domain specific terms and meanings. This paper presents four domain-specific sentiment dictionary construction methods for opinion mining, and describes performance evaluation results using a practical data set. The comparison subjects of this research include SO-PMI (Semantic Orientation from Pointwise Mutual Information) and three term frequency-based methods with different term polarity measures. To evaluate the performance of four different methods, a movie review data set from a representative Internet movie community site, IMDb (Internet Movie Database) is collected using a web crawling program, and is analyzed using R programs. Based on training data set, domain specific sentiment dictionaries are constructed using four different methods, and are compared their performance of sentiment analysis. The experimental results show that domain-specific sentiment dictionaries are working better than general-purpose dictionaries except one genre, „animation‟. Also, term frequency-based approaches show better performance than SO-PMI.

Nowadays, the audiences' consumption attitudes, consumption patterns and consumer groups are all in the great changes, thus it is necessary to improve the film’s revenue by excellent script selecting, accurate market positioning, effective product marketing, and accurate forecasting of the box office. This paper introduced the application and benefit of big data in the film industry chain in terms of film making and investing, film publicity and distribution, film broadcasting and film audience, pointed out many challenges that big data encountered in China’s film industry and finally provided useful suggestions for the practitioners in the film industry of all aspects.

Data replication is a key factor to achieve scalability and fault tolerance in databases as it maintains several clones of data objects. A change made in data automatically triggers carrying out similar changes in each of the replica. A number of data replication techniques have been proposed in the contemporary literature due to its large scale application in the real world like astronomy, high energy physics and biology. In this study we provide a critical analysis of these techniques.

This paper presents a data-driven methodology to disambiguate a query by suggesting relevant subcategories within a specific domain. This is achieved by finding correlations between the user’s search history and the context of the current search keyword. We apply automatic categorization on each query to identify a list of categories which can describe the query given. To predict the categories of a user input query, we employed machine learning algorithms. We present the preliminary evaluation results and conclude with future work.

To improve the performance of BLAS 2 GEMV subroutine under the latest instruction set, Intel AVX, this paper presents a new approach to analyze the new generation instruction set and enhance the efficiency of current data-oriented math subroutines. The whole optimizing process involves memory access optimization, SIMD optimization and parallel optimization. Also, this paper shows the comparison between the traditional SSE instruction set and the AVX instruction set. Experiments show that the optimized GEMV function has obtained considerable increase on performance. Compared with the Intel MKL, GotoBLAS, ATLAS, this optimized GEMV exceeds these BLAS implementations from 5% to 10%.

Database Reverse Engineering (DBRE) is an operation used to extract requirements from any system. The operation is implemented to facilitate the understanding of the system that has a little documentation about design and architecture. DBRE is a very important process used when database designers would like to expand the system or transition to the latest technology in DBRE fields. In the relational database model DBRE try to extract Entity Relationship Diagram (ERD) from relational database schema. Database designers find content of the data for a lot of attributes are not related with their names. In this paper, proposed methodology used to extract ERD from relational database schema with the attributes related with their names, both types of entities regular and weak entity, relationships and keys, which are found in the table that has extracted the relational database schema. The basic inputs of this approach are relational database schema that generated from database. The relational database schema used to extract the information about ERD. After that, obtain information that contain keywords help database designers to extract entities and their attributes semantics related with their names from relational database schemas. Then, Determine primary keys, foreign keys and constraints of the database system. In the final step, the ERD is successfully extracted

Massive scientific computations such as matrix, graph and network algorithms are very attractive when they come to modelling real-world data. Apache Hama is a pure BSP (Bulk Synchronous Parallel) distributed computing framework for massive scientific computations. In this paper, our experiments were conducted on a 4-node Hadoop cluster. We implement Monte Carlo algorithm of Pi in Hama and Hadoop under the same software and hardware environment. The experimental results show that Hama can achieve much higher performance than Hadoop in our testbed.

Educational data mining has received considerable attention in the last few years. Many data mining techniques are proposed to extract the hidden knowledge from educational data. The extracted knowledge helps the institutions to improve their teaching methods and learning process. All these improvements lead to enhance the performance of the students and the overall educational outputs. In this paper, we propose a new student’s performance prediction model based on data mining techniques with new data attributes/features, which are called student’s behavioral features. These type of features are related to the learner’s interactivity with the e-learning management system. The performance of student’s predictive model is evaluated by set of classifiers, namely; Artificial Neural Network, Naïve Bayesian and Decision tree. In addition, we applied ensemble methods to improve the performance of these classifiers. We used Bagging, Boosting and Random Forest (RF), which are the common ensemble methods used in the literature. The obtained results reveal that there is a strong relationship between learner’s behaviors and their academic achievement. The accuracy of the proposed model using behavioral features achieved up to 22.1% improvement comparing to the results when removing such features and it achieved up to 25.8% accuracy improvement using ensemble methods. By testing the model using newcomer students, the achieved accuracy is more than 80%. This result proves the reliability of the proposed model.