Exploiting effective features for chinese sentiment classification

Features play a fundamental role in sentiment classification. How to effectively select different types of features to improve sentiment classification performance is the primary topic of this paper. Ngram features are commonly employed in text classification tasks; in this paper, sentiment-words, substrings, substring-groups, and key-substring-groups, which have never been considered in sentiment classification area before, are also extracted as features. The extracted features are then compared and analyzed. To demonstrate generality, we use two authoritative Chinese data sets in different domains to conduct our experiments. Our statistical analysis of the experimental results indicate the following: (1) different types of features possess different discriminative capabilities in Chinese sentiment classification; (2) character bigram features perform the best among the Ngram features; (3) substring-group features have greater potential to improve the performance of sentiment classification by combining substrings of different lengths; (4) sentiment words or phrases extracted from existing sentiment lexicons are not effective for sentiment classification; (5) effective features are usually at varying lengths rather than fixed lengths.     

Measuring the degree of eyestrain caused by watching LCD and PDP devices

In this paper, we propose a new way of measuring the degree of eyestrain caused by watching LCD (Liquid Crystal Display) and PDP (Plasma Display Panel) devices. In the experiments, we used a head-mounted eye capturing device and an illumination setup that was designed to avoid specular reflections caused by glasses or contact lenses. Using the captured eye images, we analyzed the eye blinking and changes of pupil sizes (pupil accommodation), using a real-time image processing algorithm. Then we analyzed the degree of eyestrain based on the calculated blinking rate and the pupil accommodation speed. The proposed method offers five improvements over previous methods. First, we perform a comparative analysis of LCD and PDP devices based on the degree of eyestrain. Second, to analyze the degree of eyestrain, we use quantitative data such as the blinking rate and the pupil accommodation speed. Third, we measure the accurate eye blinking and changes of pupil sizes by using high-resolution and zoomed eye image sequences. Fourth, since the camera and illuminative system are based on a specular reflective model, the proposed method can be used with subjects that wear glasses or contact lenses. Fifth, the proposed method is performed at real-time speed.Experimental results showed that the degree of eyestrain when watching LCD devices was greater than that when watching PDP devices.

Relevance to industry

In the large display industry, LCD and PDP devices have become more and more prevalent. In the past, LCD devices have been compared to PDP devices in terms of factors such as spatial resolution, brightness, contrast levels, etc. To perform more accurate comparisons based on human factor, we are proposing a new way of comparing LCD to PDP devices based on the degree of eyestrain.     

Fully discrete potential-based finite element methods for a transient eddy current problem

Fully discrete potential-based finite element methods called methods are used to solve a transient eddy current problem in a three-dimensional convex bounded polyhedron. Using methods, fully discrete coupled and decoupled numerical schemes are developed. The existence and uniqueness of solutions for these schemes together with the energy-norm error estimates are provided. To verify the validity of both schemes, some computer simulations are performed for the model from TEAM Workshop Problem 7. This work was supported by Postech BSRI Research Fund-2009, National Basic Research Program of China (2008CB425701), NSFC under the grant 10671025 and the Key Project of Chinese Ministry of Education (No. 107018).     

Learning decision trees with taxonomy of propositionalized attributes

We consider the problem of exploiting a taxonomy of propositionalized attributes in order to learn compact and robust classifiers. We introduce propositionalized attribute taxonomy guided decision tree learner (PAT-DTL), an inductive learning algorithm that exploits a taxonomy of propositionalized attributes as prior knowledge to generate compact decision trees. Since taxonomies are unavailable in most domains, we also introduce propositionalized attribute taxonomy learner (PAT-Learner) that automatically constructs taxonomy from data. PAT-DTL uses top-down and bottom-up search to find a locally optimal cut that corresponds to the literals of decision rules from data and propositionalized attribute taxonomy. PAT-Learner propositionalizes attributes and hierarchically clusters the propositionalized attributes based on the distribution of class labels that co-occur with them to generate a taxonomy. Our experimental results on UCI repository data sets show that the proposed algorithms can generate a decision tree that is generally more compact than and is sometimes comparably accurate to those produced by standard decision tree learners.     

Structural reliability assessment based on probability and convex set mixed model

This paper investigates the reliability assessment of structures exhibiting both stochastic and bounded uncertainties by using a probability and convex set mixed model. The safety measure of a structure is quantified by a reliability index defined by a nested minimization problem. An iterative procedure is developed for seeking the worst-case point and the most probable failure point in the standard uncertainty space. Numerical examples are given to demonstrate the applicability of the probability and convex set mixed model representation in the structural reliability assessment, as well as to illustrate the validity and effectiveness of the proposed numerical method.     

A method of stepwise benchmarking for inefficient DMUs based on the proximity-based target selection

DEA is a useful nonparametric method of measuring the relative efficiency of a DMU and yielding a reference target for an inefficient DMU. However, it is very difficult for inefficient DMUs to be efficient by benchmarking a target DMU which has different input use. Identifying appropriate benchmarks based on the similarity of input endowment makes it easier for an inefficient DMU to imitate its target DMUs. But it is rare to find out a target DMU, which is both the most efficient and similar in input endowments, in real situation. Therefore, it is necessary to provide an optimal path to the most efficient DMU on the frontier through several times of a proximity-based target selection process. We propose a dynamic method of stepwise benchmarking for inefficient DMUs to improve their efficiency gradually.The empirical study is conducted to compare the performance between the proposed method and the prior methods with a dataset collected from Canadian Bank branches. The comparison result shows that the proposed method is very practical to obtain a gradual improvement for inefficient DMUs while it assures to reach frontier eventually.     

Fuzzy multiple goal programming applied to TFT-LCD supplier selection by downstream manufacturers

In today’s highly competitive environment, a good supply chain relationship is essential for a company to survive and to acquire reasonable profit. While a few large companies may be able to vertically integrate from the design stage to the final distribution of the entire supply chain, most companies can only focus on their specialized functions and to cooperate with upstream or downstream companies. Supplier selection, as a result, is very important for maintaining strategic alliances. The objective of this paper is to develop a fuzzy multiple goal programming (FMGP) model to help downstream companies to select thin film transistor liquid crystal display (TFT-LCD) suppliers for cooperation. Fuzzy analytic hierarchy process (FAHP) is applied first to analyze the importance of multiple factors by incorporating experts’ opinion, and these factors include cost, yield and number of suppliers. Multi-choice goal programming is used next to consider the limits of various resources and to formulate the constraints. From the experimental design and examination, we can testify that the proposed model not only can consider multi-choice goals, decision-making behavior and limit of resources, it can also allocate the purchase among the selected supplier(s).     

Scale‐aware Structure‐Preserving Texture Filtering

This paper presents a novel method to enhance the performance of structure‐preserving image and texture filtering. With conventional edge‐aware filters, it is often challenging to handle images of high complexity where features of multiple scales coexist. In particular, it is not always easy to find the right balance between removing unimportant details and protecting important features when they come in multiple sizes, shapes, and contrasts. Unlike previous approaches, we address this issue from the perspective of adaptive kernel scales. Relying on patch‐based statistics, our method identifies texture from structure and also finds an optimal per‐pixel smoothing scale. We show that the proposed mechanism helps achieve enhanced image/texture filtering performance in terms of protecting the prominent geometric structures in the image, such as edges and corners, and keeping them sharp even after significant smoothing of the original signal.