In similarity-based constrained clustering, there have been various approaches on how to define the similarity between documents to guide the grouping of similar documents together. This paper presents an approach to use term-distribution statistics extracted from a small number of cue instances with their known classes, for term weightings as indirect distance constraint. As for distribution-based term weighting, three types of term-oriented standard deviations are exploited: distribution of a term in a collection (SD), average distribution of a term in a class (ACSD), and average distribution of a term among classes (CSD). These term weightings are explored with the consideration of symmetry concepts by varying the magnitude to positive and negative for promoting and demoting effects of three standard deviations. In k-means, followed the symmetry concept, both seeded and unseeded centroid initializations are investigated and compared to the centroid-based classification. Our experiment is conducted using five English text collections and one Thai text collection, i.e., Amazon, DI, WebKB1, WebKB2, and 20Newsgroup, as well as TR, a collection of Thai reform-related opinions. Compared to the conventional TFIDF, the distribution-based term weighting improves the centroid-based method, seeded k-means, and k-means with the error reduction rate of 22.45%, 31.13%, and 58.96%.

In handwriting recognition research, a public image dataset is necessary to evaluate algorithm correctness and runtime performance. Unfortunately, in existing Thai language script image datasets, there is a lack of variety of standard handwriting types. This paper focuses on a new offline Thai handwriting image dataset named Burapha-TH. The dataset has 68 character classes, 10 digit classes, and 320 syllable classes. For constructing the dataset, 1072 Thai native speakers wrote on collection datasheets that were then digitized using a 300 dpi scanner. De-skewing, detection box and segmentation algorithms were applied to the raw scans for image extraction. The experiment used different deep convolutional models with the proposed dataset. The result shows that the VGG-13 model (with batch normalization) achieved accuracy rates of 95.00%, 98.29%, and 96.16% on character, digit, and syllable classes, respectively. The Burapha-TH dataset, unlike all other known Thai handwriting datasets, retains existing noise, the white background, and all artifacts generated by scanning. This comprehensive, raw, and more realistic dataset will be helpful for a variety of research purposes in the future.

No abstract

In image classification, various techniques have been developed to enhance the performance of principal component analysis (PCA) dimension reduction techniques with guiding weighting features to remove redundant and irrelevant features. This study proposes the statistically weighted dimension technique based on three distribution-related class behaviors; collection-class, inter-class, and intra-class to enhance the feature-extraction ability before using PCA for feature selection. The data from the statistics-weighted dimension spaces is utilized to reduce dimensionality by reducing the large index data into smaller index data using PCA. The new principal component from the weighted training part by an unlabeled dataset is constructed and then the image is classified efficiently. Additionally, the weighting direction investigates the pros and cons of promoting and demoting to determine the worst or best option utilizing the exponents of three proposed weighted scheme. The experiment is conducted using three datasets, MNIST, E-MNIST, and F-MNIST, along with three image classification algorithms, logistic Regression, KNN, and SVM (RBF). The results clearly demonstrate that the statistically weighted dimension feature can improve the conventional classification accuracy in lower dimensions with an appropriate combination of weighting nearly 3% for the best solution on dimensionality reduction by more than 50%.

No abstract