This paper demonstrates a new method for leveraging unstructured annotations to infer semantic document properties. We consider the domain of product reviews, which are often annotated by their authors with free-text keyphrases, such as “a real bargain ” or “good value. ” We leverage these unstructured annotations by clustering them into semantic properties, and then tying the induced clusters to hidden topics in the document text. This allows us to predict relevant properties of unannotated documents. Our approach is implemented in a hierarchical Bayesian model with joint inference, which increases the robustness of the keyphrase clustering and encourages document topics to correlate with semantically meaningful properties. We perform several evaluations of our model, and find that it substantially outperforms alternative approaches. 1

We show that a simple procedure based on maximizing the number of informative content-words can produce some of the best reported results for multi-document summarization. We first assign a score to each term in the document cluster, using only frequency and position information, and then we find the set of sentences in the document cluster that maximizes the sum of these scores, subject to length constraints. Our overall results are the best reported on the DUC-2004 summarization task for the ROUGE-1 score, and are the best, but not statistically significantly different from the best system in MSE-2005. Our system is also substantially simpler than the previous best system. 1

The increasing complexity of summarization systems makes it difficult to analyze exactly which modules make a difference in performance. We carried out a principled comparison between the two most commonly used schemes for assigning importance to words in the context of query focused multi-document summarization: raw frequency (word probability) and log-likelihood ratio. We demonstrate that the advantages of log-likelihood ratio come from its known distributional properties which allow for the identification of a set of words that in its entirety defines the aboutness of the input. We also find that LLR is more suitable for query-focused summarization since, unlike raw frequency, it is more sensitive to the integration of the information need defined by the user. 1

It is difficult to identify sentence importance from a single point of view. In this paper, we propose a learning-based approach to combine various sentence features. They are categorized as surface, content, relevance and event features. Surface features are related to extrinsic aspects of a sentence. Content features measure a sentence based on contentconveying words. Event features represent sentences by events they contained. Relevance features evaluate a sentence from its relatedness with other sentences. Experiments show that the combined features improved summarization performance significantly. Although the evaluation results are encouraging, supervised learning approach requires much labeled data. Therefore we investigate co-training by combining labeled and unlabeled data. Experiments show that this semisupervised learning approach achieves comparable performance to its supervised counterpart and saves about half of the labeling time cost. 1

We present a new approach to automatic summarization based on neural nets, called NetSum. We extract a set of features from each sentence that helps identify its importance in the document. We apply novel features based on news search query logs and Wikipedia entities. Using the RankNet learning algorithm, we train a pair-based sentence ranker to score every sentence in the document and identify the most important sentences. We apply our system to documents gathered from CNN.com, where each document includes highlights and an article. Our system significantly outperforms the standard baseline in the ROUGE-1 measure on over 70 % of our document set. 1

Query-oriented summarization aims at extracting an informative summary from a document collection for a given query. It is very useful to help users grasp the main information related to a query. Existing work can be mainly classified into two categories: supervised method and unsupervised method. The former requires training examples, which makes the method limited to predefined domains. While the latter usually utilizes clustering algorithms to find ‘centered ’ sentences as the summary. However, the method does not consider the query information, thus the summarization is general about the document collection itself. Moreover, most of existing work assumes that documents related to the query only talks about one topic. Unfortunately, statistics show that a large portion of summarization tasks talk about multiple topics. In this paper, we try to break limitations of the existing methods and study a new setup of the problem of multi-topic based query-oriented summarization. We propose using a probabilistic approach to solve this problem. More specifically, we propose two strategies to incorporate the query information into a probabilistic model. Experimental results on two different genres of data show that our proposed approach can effectively extract a multi-topic summary from a document collection and the summarization performance is better than baseline methods. The approach is quite general and can be applied to many other mining tasks, for example product opinion analysis and question answering. 1

Abstract—We investigate the automatic generation of topic pages as an alternative to the current Web search paradigm. Topic pages explicitly aggregate information across documents, filter redundancy, and promote diversity of topical aspects. We propose a novel framework for building rich topical aspect models and selecting diverse information from the Web. In particular, we use Web search logs to build aspect models with various degrees of specificity, and then employ these aspect models as input to a sentence selection method that identifies relevant and non-redundant sentences from the Web. Automatic and manual evaluations on biographical topics show that topic pages built by our system compare favorably to regular Web search results and to MDS-style summaries of the Web results on all metrics employed. Keywords-Web search; topic page; query log; aspect model. I.

Common approaches to assessing document quality look at shallow aspects, such as grammar and vocabulary. For many real-world applications, deeper notions of quality are needed. This work represents a first step in a project aimed at developing computational methods for deep assessment of quality in the domain of intelligence reports. We present an automated system for ranking intelligence reports with regard to coverage of relevant material. The system employs methodologies from the field of automatic summarization, and achieves performance on a par with human judges, even in the absence of the underlying information sources. 1

As the amount of information on the Web grows, the ability to retrieve relevant information quickly and easily is necessary. The combination of ample news sources on the Web, little time to browse news, and smaller mobile devices motivates the development of automatic highlight extraction from single news articles. Our system, NetSum, is the first system to produce highlights of an article and significantly outperform the baseline. Our approach uses novel information sources to exploit human interest for highlight extraction. In this paper, we briefly describe the novelties of NetSum, originally presented at EMNLP 2007, and embed our work in the AI context.

We investigate the automatic generation of topic pages as an alternative to the current Web search paradigm. We describe a general framework, which combines query log analysis to build aspect models, sentence selection methods for identifying relevant and non-redundant Web sentences, and a technique for sentence ordering. We evaluate our approach on biographical topics both automatically and manually, by using Wikipedia as reference.