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.

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.

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.

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.

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.