We have over a decade of experience in developing industrial-strength custom AI solutions. We specialize in Machine Learning & Natural Language Processing (NLP) solutions...
This paper presents a flexible framework for generating very short abstractive summaries. The key idea is to use a word graph data structure referred to as the Opinosis-Graph to represent the text to be summarized. Then, we repeatedly find paths through this graph to produce concise summaries. We consider Opinosis a "shallow" abstractive summarizer as it uses the original text itself to generate summaries. This is unlike a true abstractive summarizer that would need a deeper level of natural language understanding.
While the evaluation is on an opinion dataset, the approach itself is general in that, it can be applied to any corpus containing high amounts of redundancies, for example, Twitter comments or user comments on blog/news articles. A very similar work to ours (published at the same time and at the same conference) is the following:
Multi-sentence compression: Finding shortest paths in word graphs
Proceedings of the 23rd International Conference on Computaional Linguistics (COLING 10). Beijing, China, August 23-27, 2010. Katja Filippova
Katja's work was evaluated on a news dataset (google news) for both English and Spanish while ours was evaluated on user reviews from various sources (English only). She studies the informativeness and grammaticality of sentences and in a similar way we evaluate these aspects by studying how close the Opinosis summaries are compared to the human composed summaries in terms of information overlap and readability (using a human assessor).
Our AI strategy consulting takes the guesswork out of your AI initiatives. We co-plan your AI strategy, help with vendor selection, provide technical advise and more.
Organizations classify documents so that their text data is easier to manage and utilize. Learn how 5 companies are using document classification in practice.
Model training is a critical phase in the development of AI models. It's the process of allowing a machine learning algorithm to learn patterns based on...
This article discusses what an AI strategy means, the different types of AI strategies that you should know about, and how as a leader you can get started with an AI strategy.
The 5 side-effects of not having a big data strategy for AI. Data is the fuel for modern AI applications. Unfortunately, most companies don't have a data strategy or have an ad-hoc data strategy.
AI ethics is about releasing and implementing AI responsibly, paying attention to several considerations, from data etiquette to tool development risks, as discussed in a previous article. In this article, we’ll explore some of the ethical issues that arise with AI systems, particularly machine learning systems, when we overlook the ethical considerations of AI, often unintentionally.
K. Ganesan, C. Zhai, and J. Han. Proceedings of the 23rd International Conference on Computational Linguistics (Coling 2010), page 340--348. Beijing, China, Coling 2010 Organizing Committee, (August 2010)