Qi Han

CiteRivers: Visual Analytics of Citation Patterns

The exploration and analysis of scientific literature collections is an important task for effective knowledge management. Past interest in such document sets has spurred the development of numerous visualization approaches for their interactive analysis. They either focus on the textual content of publications, or on document metadata including authors and citations. Previously presented approaches for citation analysis aim primarily at the visualization of the structure of citation networks and their exploration. We extend the state-of-the-art by presenting an approach for the interactive visual analysis of the contents of scientific documents, and combine it with a new and flexible technique to analyze their citations. This technique facilitates user-steered aggregation of citations which are linked to the content of the citing publications using a highly interactive visualization approach. Through enriching the approach with additional interactive views of other important aspects of the data, we support the exploration of the dataset over time and enable users to analyze citation patterns, spot trends, and track long-term developments. We demonstrate the strengths of our approach through a use case and discuss it based on expert user feedback.

Hierarchy-based projection of high-dimensional labeled data to reduce visual clutter

Abstract Visualizing high-dimensional labeled data on a two-dimensional plane can quickly result in visual clutter and information overload. To address this problem, the data usually needs to be structured, so that only parts of it are displayed at a time. We present a hierarchy-based approach that projects labeled data on different levels of detail on a two-dimensional plane, whilst keeping the user׳s cognitive load between the level changes as low as possible. The approach consists of three steps: First, the data is hierarchically clustered; second, the user can determine levels of detail; third, the levels of detail are visualized one at a time on a two-dimensional plane. Animations make transitions between the levels of detail traceable, while the exploration on each level is supported by several interaction techniques, including halos, a darts view, and a magic lens. We demonstrate the applicability and usefulness of the approach with use cases from the patent domain and a question-and-answer website. In addition, we conducted a qualitative evaluation to assess the usefulness and comprehensibility of our approach.

Visual Clutter Reduction through Hierarchy-based Projection of High-dimensional Labeled Data

Visualizing high-dimensional labeled data on a two-dimensional plane can quickly result in visual clutter and information overload. To address this problem, the data usually needs to be structured, so that only parts of it are displayed at a time. We present a hierarchy-based approach that projects labeled data on different levels of detail on a two-dimensional plane, whilst keeping the users cognitive load between the level changes as low as possible. The approach consists of three steps: First, the data is hierarchically clustered; second, the user can determine levels of detail; third, the levels of detail are visualized one at a time on a two-dimensional plane. Animations make transitions between the levels of detail traceable, while the exploration on each level is supported by several interaction techniques. We demonstrate the applicability and usefulness of the approach with use cases from the patent domain and a question-and-answer website.

Tunable Discounting Mechanisms for Language Modeling

Language models are fundamental to many applications in natural language processing. Most language models are trained on training data that do not support discount parameters tuning. In this work, we present novel language models based on tunable discounting mechanisms. The language models are trained on a large training set, but their discount parameters can be tuned to a target set. We explore tunable discounting and polynomial discounting based on modified Kneser-Ney models. With the resulting implementation, our language models achieve perplexity improvements in in-domain and out-of-domain evaluation. The experimental results indicate that our new models significantly outperform the baseline model and are especially suited for domain adaptation.

Unsupervised feature adaptation for cross-domain NLP with an application to compositionality grading

In this paper, we introduce feature adaptation, an unsupervised method for cross-domain natural language processing (NLP). Feature adaptation adapts a supervised NLP system to a new domain by recomputing feature values while retaining the model and the feature definitions used on the original domain. We demonstrate the effectiveness of feature adaptation through cross-domain experiments in compositionality grading and show that it rivals supervised target domain systems when moving from generic web text to a specialized physics text domain.

Domain mining for machine translation

Massive amounts of data for data mining consist of natural language data. A challenge in natural language is to translate the data into a particular language. Machine translation can do the translation automatically. However, the models trained on data from a domain tend to perform poorly for different domains. One way to resolve this issue is to train domain adaptation translation and language models. In this work, we use visualizations to analyze the similarities of domains and explore domain detection methods by using text clustering and domain language models to discover the domain of the test data. Furthermore, we present domain adaptation language models based on tunable discounting mechanism and domain interpolation. A cross-domain evaluation of the language models is performed based on perplexity, in which considerable improvements are obtained. The performance of the domain adaptation models are also evaluated in Chinese-to-English machine translation tasks. The experimental BLEU scores indicate that the domain adaptation system significantly outperforms the baseline especially in domain adaptation scenarios.