No abstract

No abstract

In this paper, we study the aspect-based extractive summarization based on the observations that a good summary should present representative opinions on user concerned sub-aspects within limited words. According to these observations, we argue that, two requirements, i.e. representativeness and diversity, should be considered for generating a good summary in addition to the traditional requirements of aspect-relevance and sentiment intensity. We focus on the intrinsic relationship between sentences and the dependency between extracted sentences for summarization, and thus propose a novel aspect-based summarization method for online reviews, which employs an Aspect-sensitive Markov Random Walk Model to meet the representativeness requirement, as well as a greedy redundancy removal method to meet the diversity requirement. The conducted experiments verify the effectiveness of the proposed method by comparing it with the baselines which ignores representativeness and/or diversity. The experimental results also show that, the two requirements we present are both indispensable for a good summary.

Recently, Deep Convolutional Neural Networks (CNNs) have been widely applied to sentiment analysis of short texts. Naturally, word embedding techniques are used to learn continuous word representations for constructing sentence matrix as input to CNN. As for sentiment analysis of customer reviews, we argue that it is problematic to learn a single representation for a word while ignoring sentiment information and the discussed aspects. In this poster, we propose a novel word embedding model to learn sentimental word embedding given specific aspects by modeling both sentiment and syntactic context under the specific aspects. We apply our method as input to CNN for sentiment analysis in multiple domains. Experiments show that the CNN based on the proposed model can consistently achieve superior performance compared to CNN based on traditional word embedding method.