Time filter

Source Type

Mueang Nonthaburi, Thailand

Sampranpiboon P.,Rangsit University | Charnkeitkong P.,Panyapiwat Institute of Management | Feng X.,University of Waterloo
WSEAS Transactions on Environment and Development | Year: 2014

The present investigation deals with the utilization of pulp waste as an adsorbent for the removal of zinc (II) ion from aqueous solution. A series of experiments were conducted in a batch system to evaluate the performance of the pulp waste for zinc removal. The effects of pH, adsorbent dosage, initial concentration and temperature were evaluated. The optimum pH value for zinc (II) adsorption on the pulp waste was found to be pH 6.0. The equilibrium sorption data were analyzed using Freundlich, Langmuir, Temkin, Halsay, Hurkins- Jura, Redlich-Peterson, Dubinin-Radushkevich and Jovanovich isotherm models, and the Langmuir and Temkin models were found to be adequate in describing the zinc (II) adsorption onto the pulp waste. Source

Sanguansat P.,Panyapiwat Institute of Management
ECTI-CON 2015 - 2015 12th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology | Year: 2015

In survey research, the offline paper-based questionnaire still necessary for data collection. Although the online survey is more convenient, internet system and portable device are needed for operating. That is not practical for some surveys. Furthermore, the paper-based questionnaire requires data entry by human which will take long time and easy to mistake. Therefore, this paper proposes the automated data entry by Optical Mark Recognition (OMR) with the proposed paper-based questionnaire. This questionnaire will input by scanner and the data will be displayed in the appropriate report for data analysis. However, the pattern of questionnaires is different for each survey. In this paper, the method for creating the questionnaire, which contains only closed-ended questions, is also proposed for user self-designed including the proper report of the output that will be opened or edited in the well-known spreadsheet software. According to the experimental results, the accuracy rate is high and suitable for real application that is the average of accuracy rate is 93.36 % for choice selection by three patterns of the markers. © 2015 IEEE. Source

Mutchima P.,Suan Dusit Rajabhat University | Sanguansat P.,Panyapiwat Institute of Management
2012 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2012 | Year: 2012

Determination of content importance is very important in achieving high quality classification. Term weighting schemes in text classification will be applied to classify videos by measuring importance of video contents. In other words, a video sequence can be treated as a document, and frames of a video are considered as words or terms which identify contents of a video. And to enhance the efficiency of video classification, this paper proposes a novel term weighting scheme, called the Term Frequency - Relevance and Non-relevance Frequency (TF-RNF) weighting. This technique can filter both relevant and non-relevant contents so as to reduce classification errors. Empirical evaluations of results show that the proposed technique significantly outperforms traditional techniques in sports video classification. © 2012 IEEE. Source

Sanguansat P.,Panyapiwat Institute of Management
2016 8th International Conference on Knowledge and Smart Technology, KST 2016 | Year: 2016

This paper proposes the sentiment analysis system in Thai language. It aims to use for the three business types (Retail, Banking and Telecommunication) to monitor their brand image via social media. Pantip.com is the most popular online community in Thailand, which many customers posted the comments about their business. Normally, three sentiments must be identified (positive, negative and neutral), but four sentiments (positive, negative, neutral and need) are introduced in our proposed system because the need sentiment can be used for generating new business opportunities. The unsupervised deep learning feature extraction for text, called Paragraph2Vec, paragraph vector or Doc2Vec, was applied in this paper, compared to the classical TF-IDF. The experimental results show that our proposed method perform better than the baseline method. © 2016 IEEE. Source

Phiwma N.,Suan Dusit Rajabhat University | Sanguansat P.,Panyapiwat Institute of Management
International Arab Journal of Information Technology | Year: 2014

In this paper, we propose new methods for feature extraction and soft majority voting to adjust efficiency and accuracy of music retrieval. For our work, the input is humming sound which is sound wave and Musical Instrument Digital Interface (MIDI) is used as the reference song in database. A critical issue of humming sound are variation such as duration, sound, tempo, key, and noise interference from both environment and acquisition instruments. Besides all the problems of humming sound we have mentioned earlier, whether humming sound and MIDI in different domain which will make the difficulty for two domains to compare each other. However, to make these two in the same domain, we convert them into the frequency domain. Our approach starts from pre-processing by using features for note segmentation by humming sound. The process consists of four steps as follows: Firstly, the MIDI is already a sequence of pitch while the pitch in humming sound is needed to extract by Subharmonic-to-Harmonic (SHR). Subsequently, the extracted pitch can be used to calculate all above attributes and then multiple classifiers are applied to classify the multiple subsets of these features. Afterwards, the subset contain the multiple attributes, Multi-Dimensional Dynamic Time Warping (MD-DTW) is used for similarity measurement. Finally, Nearest Neighbours (NN) and soft majority voting are used to obtain the retrieval results in case of equal scores. From the experiments, to achieve 100% accuracy rate at the early top-n rank in retrieving, the appropriate feature set should consist of five classifiers. Source

Discover hidden collaborations