2018

Exploiting multiple word embeddings and one-hot character vectors for aspect-based sentiment analysis

International Journal of Approximate Reasoning (ISSN: 0888-613X) Bài báo Tạp chí SCI

Research

Representing words as real-value vectors and use them as input in deep neural networks is an effective approach in many natural language processing tasks. Currently, some studies use a lower-level representation which is character-based vectors. This paper addresses on how to integrate different representations of input for the problem of aspect-based sentiment analysis. We will propose a joint model of multiple Convolutional Neural Networks (CNNs) in which each individual representation of the input is handled by one CNN. In this work we focus on three kinds of representation including word embeddings from the two methods (Word2Vec and GloVe) and the one-hot character vectors. Our experimental results demonstrate that the proposed model can achieve state-of the-art performance in aspect category detection and aspect sentiment classification tasks.

- Tác giả: Phạm Đức Hồng, Lê Anh Cường
- Trạng thái: Đã xuất bản
2018

An efficient image retrieval method using adaptive weights

Applied Intelligence (ISSN: 0924-669X) Bài báo Tạp chí SCI

Research

In existing image retrieval systems using a multipoint query, the neighborhood of the various optimal query points is determined in the same way. The different nature of the various optimal query points was not taken into consideration. This approach confines the performance of the system. To overcome this confinement, in this paper, we propose an image retrieval method through adaptive weights (Aweight) which is possible to compute the various optimal query points, optimal weights and improved distance functions to improve accuracy. In addition, our method constructs clusters without re-clustering the whole feedback image set. The experiments were performed on a set of 10,800 images and the results demonstrate that the proposed method improves performance of system in terms of accuracy.

- Tác giả: Nguyễn Hữu Quỳnh, Đào Thị Thúy Quỳnh, Phương Văn Cảnh, Nguyễn Văn Căn, Ngô Quốc Tạo
- Trạng thái: Đã xuất bản
- Fulltext: Chưa có
2017

Increasing efficiency of photovoltaic systems under non-homogeneous solar irradiation using improved Dynamic Programming methods

Solar Energy (ISSN: 0038-092X) Bài báo Tạp chí SCI

Research

The paper presents a complete technique, based on the combination of algorithms, devoted to minimize losses and increase efficiency of Total Cross Tied (TCT) connected photovoltaic (PV) systems under non- homogeneous solar irradiation, based on irradiance equalization criterion. Irradiance equalization is achieved by changing the connections of the solar panels adaptively by a dynamic switching matrix so that total solar radiation on parallel circuits is the most equalized.

In this paper, the authors introduce two algorithms. The first one is SmartChoice (SC) algorithm, which is combined with Dynamic Programming (DP) in order to create a hybrid method and obtain better results as compared to established methods for irradiance equalization. The second one is the control algorithm improvement method from Munkres’ Assignment Algorithm (MAA) that helps to increase pro- cessing speed and lengthen the lifetime of the solar power system by 56% compared with the older MAA.

By emulating and experiencing the operation of the PV system under non-homogeneous irradiation condition, obtained results show efficiency and benefits of the proposed method applied to the solar power system operation while lengthening the lifetime of the switching matrix.

- Tác giả: Ngô Ngọc Thành, Nguyễn Phùng Quang, Nguyễn Tùng Linh, Eleonora Riva Sanseverino, Pietro Romano, Fabio Viola
- Trạng thái: Đã xuất bản
2015

Dynamic programming and Munkres algorithm for optimal photovoltaic arrays reconfiguration

Solar Energy (ISSN: 0038-092X) Bài báo Tạp chí SCI

Research

In this paper, an original formulation of the control problem for optimal PV array reconfiguration, following a Total Cross Tied layout, is proposed. The formulation follows the well-known subset sum problem, which is a special case of the knapsack problem. The reconfiguration is a measure devoted to mitigate the mismatch effect and maximize the output power of small photovoltaic plants under non-homogeneous working conditions. Therefore, reconfiguration means changing the connections of the solar panels adaptively by a dynamic switching matrix. The control system implements an easy dynamic programming algorithm to change the switches layout.
The use of the Munkres assignment method in a post-processing module makes the algorithm able to obtain the optimum configuration for which it is possible to balance and minimize the aging of the switches within the switching matrix.

- Tác giả: Eleonora Riva Sanseverino, Ngô Ngọc Thành, Marzia Cardinale, Vincenzo Li Vigni, Domenico Musso, Pietro Romano, Fabio Viola
- Trạng thái: Đã xuất bản
2019

Multichannel image contrast enhancement based on linguistic rule-based intensificators

Applied Soft Computing Journal (ISSN: 1568-4946) Bài báo Tạp chí SCIE

Research

- Tác giả: Ngô Hoàng Huy, Nguyễn Cát Hồ, Nguyễn Văn Quyền
- Trạng thái: Đã xuất bản
2018

Learning multiple layers of knowledge representation for aspect based sentiment analysis

Data & Knowledge Engineering (ISSN: 0169-023X) Bài báo Tạp chí SCIE

Research

Sentiment Analysis is the task of automatically discovering the exact sentimental ideas about a product (or service, social event, etc.) from customer textual comments (i.e. reviews) crawled from various social media resources. Recently, we can see the rising demand of aspect-based sentiment analysis, in which we need to determine sentiment ratings and importance degrees of product aspects. In this paper we propose a novel multi-layer architecture for representing customer reviews. We observe that the overall sentiment for a product is composed from sentiments of its aspects, and in turn each aspect has its sentiments expressed in related sentences which are also the compositions from their words. This observation motivates us to design a multiple layer architecture of knowledge representation for representing the different sentiment levels for an input text. This representation is then integrated into a neural network to form a model for prediction of product overall ratings. We will use the representation learning techniques including word embeddings and compositional vector models, and apply a back-propagation algorithm based on gradient descent to learn the model. This model consequently generates the aspect ratings as well as aspect weights (i.e. aspect importance degrees). Our experiment is conducted on a data set of reviews from hotel domain, and the obtained results show that our model outperforms the well-known methods in previous studies.

- Tác giả: Phạm Đức Hồng, Lê Anh Cường
- Trạng thái: Đã xuất bản
2017

Bisimilarity for paraconsistent description logics

Journal of Intelligent & Fuzzy Systems (ISSN: 1064-1246) Bài báo Tạp chí SCIE

Research

- Tác giả: Nguyễn Anh Linh, Nguyễn Thị Hồng Khánh, Hà Quang Thụy, Nguyễn Ngọc Thành
- Trạng thái: Đã xuất bản
- Fulltext: Chưa có
2017

An efficient semantic – Related image retrieval method

Expert Systems with Applications (ISSN: 0957-4174) Bài báo Tạp chí SCIE

Research

Many previous techniques were designed to retrieve semantic images in a particular neighborhood of the image query and thus bypassing the semantically related images in the whole feature space. Several methods were originally designed to retrieve semantically images in the same space feature but with low precision. In this paper, we propose a semantic - Related Image Retrieval method (SRIR), which can retrieve semantic images of the same space in high precision. Our method takes advantage of the user feedback to determine the semantic importance of each query and the importance of each feature. In addition, the retrieval time of our method does not increase with the number of user feedback. We also provide experimental results to demonstrate the effectiveness of our method.

- Tác giả: Đào Thị Thúy Quỳnh, Nguyễn Hữu Quỳnh, Phương Văn Cảnh, Ngô Quốc Tạo
- Trạng thái: Đã xuất bản
2012

An efficient content based image retrieval method for retrieving images

International journal of innovative computing, information and control (ISSN: 1349-4198) Bài báo Tạp chí SCIE

Research

Quality, efficiency and scalability are the key issues in the design of image retrieval systems for large image databases. Although the quality of image retrieval methods still depends strongly on the application domain, color based retrieval techniques have been shown to be competitive and generally applicable. However, the retrieval algorithms based on color histograms largely ignore spatial information in the matching process. In this paper, we propose a technique to improve the retrieval process by image regions matching. We carried out an experiment on an image database containing 8000 images. The experimental results show that our proposed technique is more effective than the other retrieval techniques such as color histogram based and Color Based Clustering based techniques.

- Trạng thái: Đã xuất bản
- Fulltext: Chưa có
2010

A novel content based image retrieval method based on splitting the image into homogeneous regions

International journal of innovative computing, information and control (ISSN: 1349-4198) Bài báo Tạp chí SCIE

Research

Color histograms are widely used for content-based image retrieval. Their advantages are efficiency and insensitivity to small changes in camera viewpoint. However, this technique does not employ spatial information of image colors. To take spatial information into account, some local color feature-based approaches proposed. However, these approaches have some disadvantages such as low effectiveness, consuming computation times and storage. This paper proposes a novel technique for retrieving images called HR (Homogeneous Region), which is based on splitting images into two homogeneous regions. We carry out an experiment on an image database containing 10,000 images. Experimental results show that our technique is more effective than the color coherence vector (CCV) and Color Based Clustering based retrieval techniques.

- Trạng thái: Đã xuất bản
- Fulltext: Chưa có