Vision–Language Models (VLMs) have shown strong performance on chart understanding, yet remain brittle to misleading visual encodings—cases where the underlying data and question stay the same, but design choices (e.g., inverted or truncated axes) induce a misleading visual impression. Prior work shows that supervising VLMs with human gaze improves attention alignment and accuracy, but it is unknown whether gaze also improves robustness to deceptive chart designs. We compare gaze-supervised and non-gaze-supervised VLMs under a set of controlled chart transformations that preserve question semantics while altering visual encodings, including axis inversions, truncations, and chart-type mismatches. Across manipulation categories, gaze-supervised models exhibit smaller performance fluctuations and fewer statistically significant prediction shifts than their non-gaze counterparts, particularly under transformations that alter temporal or axis encodings. These results suggest that learning where humans look can strengthen attention alignment and improve robustness to deceptive visualizations, underscoring the value of human-centric supervision.

Md Tahmid Rahman Laskar, Israt Jahan, Elham Dolatabadi, Chun Peng, Enamul Hoque, Jimmy Huang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Ridwan Mahbub, Mohammed Saidul Islam, Mir Tafseer Nayeem, Md Tahmid Rahman Laskar, Mizanur Rahman, Shafiq Joty, Enamul Hoque. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Automated data visualization plays a crucial role in simplifying data interpretation, enhancing decision-making, and improving efficiency.While large language models (LLMs) have shown promise in generating visualizations from natural language, the absence of comprehensive benchmarks limits the rigorous evaluation of their capabilities.We introduce Text2Vis, a benchmark designed to assess textto-visualization models, covering 20+ chart types and diverse data science queries, including trend analysis, correlation, outlier detection, and predictive analytics.It comprises 1,985 samples, each with a data table, natural language query, short answer, visualization code, and annotated charts.The queries involve complex reasoning, conversational turns, and dynamic data retrieval.We benchmark 11 open-source and closed-source models, revealing significant performance gaps, highlighting key challenges, and offering insights for future advancements.To close this gap, we propose the first cross-modal actor-critic agentic framework that jointly refines the textual answer and visualization code, increasing GPT-4o's pass rate from 26% to 42% over the direct approach and improving chart quality.We also introduce an automated LLM-based evaluation framework that enables scalable assessment across thousands of samples without human annotation, measuring answer correctness, code execution success, visualization readability, and chart accuracy.We release Text2Vis at https: //github.com/vis-nlp/Text2Vis.

Aligning visual features with language embeddings is a key challenge in vision-language models (VLMs). The performance of such models hinges on having a good connector that maps visual features generated by a vision encoder to a shared embedding space with the LLM while preserving semantic similarity. Existing connectors, such as multilayer perceptrons (MLPs), lack inductive bias to constrain visual features within the linguistic structure of the LLM’s embedding space, making them data-hungry and prone to cross-modal misalignment. In this work, we propose a novel vision-text alignment method, AlignVLM, that maps visual features to a weighted average of LLM text embeddings. Our approach leverages the linguistic priors encoded by the LLM to ensure that visual features are mapped to regions of the space that the LLM can effectively interpret. AlignVLM is particularly effective for document understanding tasks, where visual and textual modalities are highly correlated. Our extensive experiments show that AlignVLM achieves state-of-the-art performance compared to prior alignment methods, with larger gains on document understanding and under low-resource setups. We provide further analysis demonstrating its efficiency and robustness to noise.

Abstract Natural language and visualization are two complementary modalities of human communication that play a crucial role in conveying information effectively. While visualizations help people discover trends, patterns and anomalies in data, natural language descriptions help explain these insights. Thus, combining text with visualizations is a prevalent technique for effectively delivering the core message of the data. Given the rise of natural language generation (NLG), there is a growing interest in automatically creating natural language descriptions for visualizations, which can be used as chart captions, answering questions about charts or telling data‐driven stories. In this survey, we systematically review the state of the art on NLG for visualizations and introduce a taxonomy of the problem. The NLG tasks fall within the domain of natural language interfaces (NLIs) for visualization, an area that has garnered significant attention from both the research community and industry. To narrow down the scope of the survey, we primarily concentrate on the research works that focus on text generation for visualizations. To characterize the NLG problem and the design space of proposed solutions, we pose five Wh‐questions, why and how NLG tasks are performed for visualizations, what the task inputs and outputs are, as well as where and when the generated texts are integrated with visualizations. We categorize the solutions used in the surveyed papers based on these ‘five Wh‐questions’. Finally, we discuss the key challenges and potential avenues for future research in this domain.

Large Language Models (LLMs) have emerged as one of the most important breakthroughs in natural language processing (NLP) for their impressive skills in language generation and other language-specific tasks. Though LLMs have been evaluated in various tasks, mostly in English, they have not yet undergone thorough evaluation in under-resourced languages such as Bengali (Bangla). To this end, this paper introduces BenLLM-Eval, which consists of a comprehensive evaluation of LLMs to benchmark their performance in the low-resourced Bangla language. In this regard, we select various important and diverse Bangla NLP tasks, such as text summarization, question answering, paraphrasing, natural language inference, text classification, and sentiment analysis for zero-shot evaluation of popular LLMs, namely, ChatGPT, LLaMA-2, and Claude-2. Our experimental results demonstrate that while in some Bangla NLP tasks, zero-shot LLMs could achieve performance on par, or even better than current SOTA fine-tuned models; in most tasks, their performance is quite poor (with the performance of open-source LLMs like LLaMA-2 being significantly bad) in comparison to the current SOTA results. Therefore, it calls for further efforts to develop a better understanding of LLMs in low-resource languages like Bangla.

Data-driven storytelling is a powerful method for conveying insights by combining narrative techniques with visualizations and text. These stories integrate visual aids, such as highlighted bars and lines in charts, along with textual annotations explaining insights. However, creating such stories requires a deep understanding of the data and meticulous narrative planning, often necessitating human intervention, which can be time-consuming and mentally taxing. While Large Language Models (LLMs) excel in various NLP tasks, their ability to generate coherent and comprehensive data stories remains underexplored. In this work, we introduce a novel task for data story generation and a benchmark containing 1,449 stories from diverse sources. To address the challenges of crafting coherent data stories, we propose a multi-agent framework employing two LLM agents designed to replicate the human storytelling process: one for understanding and describing the data (Reflection), generating the outline, and narration, and another for verification at each intermediary step. While our agentic framework generally outperforms non-agentic counterparts in both model-based and human evaluations, the results also reveal unique challenges in data story generation.

Md Tahmid Rahman Laskar, Sawsan Alqahtani, M Saiful Bari, Mizanur Rahman, Mohammad Abdullah Matin Khan, Haidar Khan, Israt Jahan, Amran Bhuiyan, Chee Wei Tan, Md Rizwan Parvez, Enamul Hoque, Shafiq Joty, Jimmy Huang. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Charts are widely used for data analysis, providing visual representations and insights into complex data.To facilitate chart-based data analysis using natural language, several downstream tasks have been introduced recently such as chart question answering and chart summarization.However, existing methods for these tasks often rely on pretraining on language or vision-language tasks, neglecting the explicit modeling of chart structures (e.g., how chart elements are related to each other).To address this, we first build a large corpus of charts covering diverse topics and visual styles.We then present UniChart, a pretrained model for chart comprehension and reasoning.UniChart encodes the relevant text, data, and visual elements of charts and then uses a chart-grounded text decoder for text generation.We propose several chart-specific pretraining tasks that include: (i) low-level tasks to extract the visual elements (e.g., bars, lines) and data from charts, and (ii) high-level tasks to acquire chart understanding and reasoning skills.Our experiments demonstrate that pretraining UniChart on a large corpus with chart-specific objectives, followed by fine-tuning, yields state-of-the-art performance on four downstream tasks.Moreover, our model exhibits superior generalizability to unseen chart corpus, surpassing previous approaches that lack chart-specific objectives and utilize limited chart resources.

Web-based data visualizations have become very popular for exploring data and communicating insights. Newspapers, journals, and reports regularly publish visualizations to tell compelling stories with data. Unfortunately, most visualizations are inaccessible to readers with visual impairments. For many charts on the web, there are no accompanying alternative (alt) texts, and even if such texts exist they do not adequately describe important insights from charts. To address the problem, we first interviewed 15 blind users to understand their challenges and requirements for reading data visualizations. Based on the insights from these interviews, we developed SeeChart, an interactive tool that automatically deconstructs charts from web pages and then converts them to accessible visualizations for blind people by enabling them to hear the chart summary as well as to interact through data points using the keyboard. Our evaluation with 14 blind participants suggests the efficacy of SeeChart in understanding key insights from charts and fulfilling their information needs while reducing their required time and cognitive burden.

Shankar Kantharaj, Rixie Tiffany Leong, Xiang Lin, Ahmed Masry, Megh Thakkar, Enamul Hoque, Shafiq Joty. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Charts are very popular to analyze data and convey important insights. People often analyze visualizations to answer open-ended questions that require explanatory answers. Answering such questions are often difficult and time-consuming as it requires a lot of cognitive and perceptual efforts. To address this challenge, we introduce a new task called OpenCQA, where the goal is to answer an open-ended question about a chart with descriptive texts. We present the annotation process and an in-depth analysis of our dataset. We implement and evaluate a set of baselines under three practical settings. In the first setting, a chart and the accompanying article is provided as input to the model. The second setting provides only the relevant paragraph(s) to the chart instead of the entire article, whereas the third setting requires the model to generate an answer solely based on the chart. Our analysis of the results show that the top performing models generally produce fluent and coherent text while they struggle to perform complex logical and arithmetic reasoning.

Abstract Information visualizations such as bar charts and line charts are very common for analyzing data and discovering critical insights. Often people analyze charts to answer questions that they have in mind. Answering such questions can be challenging as they often require a significant amount of perceptual and cognitive effort. Chart Question Answering (CQA) systems typically take a chart and a natural language question as input and automatically generate the answer to facilitate visual data analysis. Over the last few years, there has been a growing body of literature on the task of CQA. In this survey, we systematically review the current state‐of‐the‐art research focusing on the problem of chart question answering. We provide a taxonomy by identifying several important dimensions of the problem domain including possible inputs and outputs of the task and discuss the advantages and limitations of proposed solutions. We then summarize various evaluation techniques used in the surveyed papers. Finally, we outline the open challenges and future research opportunities related to chart question answering.

Abstract The Query-Focused Text Summarization (QFTS) task aims at building systems that generate the summary of the text document(s) based on the given query. A key challenge in addressing this task is the lack of large labeled data for training the summarization model. In this article, we address this challenge by exploring a series of domain adaptation techniques. Given the recent success of pre-trained transformer models in a wide range of natural language processing tasks, we utilize such models to generate abstractive summaries for the QFTS task for both single-document and multi-document scenarios. For domain adaptation, we apply a variety of techniques using pre-trained transformer-based summarization models including transfer learning, weakly supervised learning, and distant supervision. Extensive experiments on six datasets show that our proposed approach is very effective in generating abstractive summaries for the QFTS task while setting a new state-of-the-art result in several datasets across a set of automatic and human evaluation metrics.

Increased access to online engagement platforms has created a shift in civic practice, enabling civic leaders to broaden their outreach to collect a larger number of community input, such as comments and ideas. However, sensemaking of such input remains a challenge due to the unstructured nature of text comments and ambiguity of human language. Hence, community input is often left unanalyzed and unutilized in policymaking. To address this problem, we interviewed 14 civic leaders to understand their practices and requirements. We identified challenges around organizing the unstructured community input and surfacing community’s reflections beyond binary sentiments. Based on these insights, we built CommunityPulse, an interactive system that combines text analysis and visualization to scaffold different facets of community input. Our evaluation with another 15 experts suggests CommunityPulse’s efficacy in surfacing multiple facets such as reflections, priorities, and hidden insights while reducing the required time, effort, and expertise for community input analysis.

People often use charts to analyze data, answer questions and explain their answers to others. In a formative study, we find that such human-generated questions and explanations commonly refer to visual features of charts. Based on this study, we developed an automatic chart question answering pipeline that generates visual explanations describing how the answer was obtained. Our pipeline first extracts the data and visual encodings from an input Vega-Lite chart. Then, given a natural language question about the chart, it transforms references to visual attributes into references to the data. It next applies a state-of-the-art machine learning algorithm to answer the transformed question. Finally, it uses a template-based approach to explain in natural language how the answer is determined from the chart's visual features. A user study finds that our pipeline-generated visual explanations significantly outperform in transparency and are comparable in usefulness and trust to human-generated explanations.

In the Query Focused Multi-Document Summarization (QF-MDS) task, a set of documents and a query are given where the goal is to generate a summary from these documents based on the given query.However, one major challenge for this task is the lack of availability of labeled training datasets.To overcome this issue, in this paper, we propose a novel weakly supervised learning approach via utilizing distant supervision.In particular, we use datasets similar to the target dataset as the training data where we leverage pre-trained sentence similarity models to generate the weak reference summary of each individual document in a document set from the multidocument gold reference summaries.Then, we iteratively train our summarization model on each single-document to alleviate the computational complexity issue that occurs while training neural summarization models in multiple documents (i.e., long sequences) at once.Experimental results on the Document Understanding Conferences (DUC) datasets show that our proposed approach sets a new state-of-the-art result in terms of various evaluation metrics.

Information visualizations such as bar charts and line charts are very popular for exploring data and communicating insights.Interpreting and making sense of such visualizations can be challenging for some people, such as those who are visually impaired or have low visualization literacy.In this work, we introduce a new dataset and present a neural model for automatically generating natural language summaries for charts.The generated summaries provide an interpretation of the chart and convey the key insights found within that chart.Our neural model is developed by extending the state-of-the-art model for the data-to-text generation task, which utilizes a transformerbased encoder-decoder architecture.We found that our approach outperforms the base model on a content selection metric by a wide margin (55.42% vs. 8.49%) and generates more informative, concise, and coherent summaries.

Typically, people perform visual data analysis using mouse and touch interactions. While such interactions are often easy to use, they can be inadequate for users to express complex information and may require many steps to complete a task. Recently natural language interaction has emerged as a promising technique for supporting exploration with visualization, as the user can express a complex analytical question more easily. In this paper, we investigate how to synergistically combine language and mouse-based direct manipulations so that weakness of one modality can be complemented by the other. To this end, we have developed a novel system, named Multimodal Interactions System for Visual Analysis (MIVA), that allows user to provide input using both natural language (e.g., through speech) and direct manipulation (e.g., through mouse or touch) and presents the answer accordingly. To answer the current question in the context of past interactions, the system incorporates previous utterances and direct manipulations made by the user within a finite-state model. We tested the applicability of MIVA on several dashboards including a COVID-19 dashboard that visualizes coronavirus cases around the globe. Our demonstration provides initial indication that the MIVA system enhances the flow of visual analysis by enabling fluid, iterative exploration and refinement of data in a dashboard with multiple-coordinated views.

There has been an exponential growth of asynchronous online conversations thanks to the rise of social media. Analyzing and gaining insights from such conversations can be quite challenging for a user, especially when the discussion becomes very long. Traditional sites present a conversation in a paginated list view, making it very difficult to find comments of interests about a specific topic and/or opinions which may be scattered around a long thread of discussion. In this paper, we introduce a natural language interface that supports the user to quickly locate and browse through the comments that are relevant to her information needs. Our system takes a question asked by the reader about a conversation as input and then automatically finds the answer using natural language processing techniques. It then presents the results by highlighting in a visual interface, enabling the user to quickly navigate through the comments that match her information needs. Our case studies with three users suggest that the system can help the user to effectively fulfill her information needs by highlighting the relevant comments to their question.

Word embeddings that consider context have attracted great attention for various natural language processing tasks in recent years. In this paper, we utilize contextualized word embeddings with the transformer encoder for sentence similarity modeling in the answer selection task. We present two different approaches (feature-based and fine-tuning-based) for answer selection. In the feature-based approach, we utilize two types of contextualized embeddings, namely the Embeddings from Language Models (ELMo) and the Bidirectional Encoder Representations from Transformers (BERT) and integrate each of them with the transformer encoder. We find that integrating these contextual embeddings with the transformer encoder is effective to improve the performance of sentence similarity modeling. In the second approach, we fine-tune two pre-trained transformer encoder models for the answer selection task. Based on our experiments on six datasets, we find that the fine-tuning approach outperforms the feature-based approach on all of them. Among our fine-tuning-based models, the Robustly Optimized BERT Pretraining Approach (RoBERTa) model results in new state-of-the-art performance across five datasets.

Natural language interaction has evolved as a useful modality to help users explore and interact with their data during visual analysis. Little work has been done to explore how autocompletion can help with data discovery while helping users formulate analytical questions. We developed a system called \system as a design probe to better understand the usefulness of autocompletion for visual analysis. We ran three Mechanical Turk studies to evaluate user preferences for various text- and visualization widget-based autocompletion design variants for helping with partial search queries. Our findings indicate that users found data previews to be useful in the suggestions. Widgets were preferred for previewing temporal, geospatial, and numerical data while text autocompletion was preferred for categorical and hierarchical data. We conducted an exploratory analysis of our system implementing this specific subset of preferred autocompletion variants. Our insights regarding the efficacy of these autocompletion suggestions can inform the future design of natural language interfaces supporting visual analysis.

We present a search engine for D3 visualizations that allows queries based on their visual style and underlying structure. To build the engine we crawl a collection of 7860 D3 visualizations from the Web and deconstruct each one to recover its data, its data-encoding marks and the encodings describing how the data is mapped to visual attributes of the marks. We also extract axes and other non-data-encoding attributes of marks (e.g., typeface, background color). Our search engine indexes this style and structure information as well as metadata about the webpage containing the chart. We show how visualization developers can search the collection to find visualizations that exhibit specific design characteristics and thereby explore the space of possible designs. We also demonstrate how researchers can use the search engine to identify commonly used visual design patterns and we perform such a demographic design analysis across our collection of D3 charts. A user study reveals that visualization developers found our style and structure based search engine to be significantly more useful and satisfying for finding different designs of D3 charts, than a baseline search engine that only allows keyword search over the webpage containing a chart.

Document authors commonly use tables to support arguments presented in the text. But, because tables are usually separate from the main body text, readers must split their attention between different parts of the document. We present an interactive document reader that automatically links document text with corresponding table cells. Readers can select a sentence (or tables cells) and our reader highlights the relevant table cells (or sentences). We provide an automatic pipeline for extracting such references between sentence text and table cells for existing PDF documents that combines structural analysis of tables with natural language processing and rule-based matching. On a test corpus of 330 (sentence, table) pairs, our pipeline correctly extracts 48.8% of the references. An additional 30.5% contain only false negatives (FN) errors -- the reference is missing table cells. The remaining 20.7% contain false positives (FP) errors -- the reference includes extraneous table cells and could therefore mislead readers. A user study finds that despite such errors, our interactive document reader helps readers match sentences with corresponding table cells more accurately and quickly than a baseline document reader.

Interactive visual data analysis is most productive when users can focus on answering the questions they have about their data, rather than focusing on how to operate the interface to the analysis tool. One viable approach to engaging users in interactive conversations with their data is a natural language interface to visualizations. These interfaces have the potential to be both more expressive and more accessible than other interaction paradigms. We explore how principles from language pragmatics can be applied to the flow of visual analytical conversations, using natural language as an input modality. We evaluate the effectiveness of pragmatics support in our system Evizeon, and present design considerations for conversation interfaces to visual analytics tools.

Community question answering (CQA) forums can provide effective means for sharing information and addressing a user's information needs about particular topics. However, many such online forums are not moderated, resulting in many low quality and redundant comments, which makes it very challenging for users to find the appropriate answers to their questions. In this paper, we apply a user-centered design approach to develop a system, CQAVis, which supports users in identifying high quality comments and get their questions answered. Informed by the user's requirements, the system combines both text analytics and interactive visualization techniques together in a synergistic way. Given a new question posed by the user, the text analytic module automatically finds relevant answers by exploring existing related questions and the comments within their threads. Then the visualization module presents the search results to the user and supports the exploration of related comments. We have evaluated the system in the wild by deploying it within a CQA forum among thousands of real users. Through the online study, we gained deeper insights about the potential utility of the system, as well as learned generalizable lessons for designing visual text analytics systems for the domain of CQA forums.

This paper addresses the problem of speech act recognition in written asynchronous conversations (e.g., fora, emails).We propose a class of conditional structured models defined over arbitrary graph structures to capture the conversational dependencies between sentences.Our models use sentence representations encoded by a long short term memory (LSTM) recurrent neural model.Empirical evaluation shows the effectiveness of our approach over existing ones: (i) LSTMs provide better task-specific representations, and (ii) the global joint model improves over local models.

Online conversations, such as blogs, provide rich amount of information and opinions about popular queries. Given a query, traditional blog sites return a set of conversations often consisting of thousands of comments with complex thread structure. Since the interfaces of these blog sites do not provide any overview of the data, it becomes very difficult for the user to explore and analyze such a large amount of conversational data. In this paper, we present MultiConVis, a visual text analytics system designed to support the exploration of a collection of online conversations. Our system tightly integrates NLP techniques for topic modeling and sentiment analysis with information visualizations, by considering the unique characteristics of online conversations. The resulting interface supports the user exploration, starting from a possibly large set of conversations, then narrowing down to the subset of conversations, and eventually drilling-down to the set of comments of one conversation. Our evaluations through case studies with domain experts and a formal user study with regular blog readers illustrate the potential benefits of our approach, when compared to a traditional blog reading interface.

Since the mid-2000s, there has been exponential growth of asynchronous online conversations, thanks to the rise of social media. Analyzing and gaining insights from such conversations can be quite challenging for a user, especially when the discussion becomes very long. A promising solution to this problem is topic modeling, since it may help the user to understand quickly what was discussed in a long conversation and to explore the comments of interest. However, the results of topic modeling can be noisy, and they may not match the user’s current information needs. To address this problem, we propose a novel topic modeling system for asynchronous conversations that revises the model on the fly on the basis of users’ feedback. We then integrate this system with interactive visualization techniques to support the user in exploring long conversations, as well as in revising the topic model when the current results are not adequate to fulfill the user’s information needs. Finally, we report on an evaluation with real users that compared the resulting system with both a traditional interface and an interactive visual interface that does not support human-in-the-loop topic modeling. Both the quantitative results and the subjective feedback from the participants illustrate the potential benefits of our interactive topic modeling approach for exploring conversations, relative to its counterparts.

In the last decade, there has been an exponential growth of asynchronous online conversations thanks to the rise of social media. Analyzing and gaining insights from such conversations can be quite challenging for a user, especially when the discussion becomes very long. A promising solution to this problem is topic modeling, since it may help the user to quickly understand what was discussed in the long conversation and explore the comments of interest. However, the results of topic modeling can be noisy and may not match the user's current information needs. To address this problem, we propose a novel topic modeling system for asynchronous conversations that revises the model on the fly based on user's feedback. We then integrate this system with interactive visualization techniques to support the user in exploring long conversations, as well as revising the topic model when the current results are not adequate to fulfill her information needs. An evaluation with real users illustrates the potential benefits of our approach for exploring conversations, when compared to both a traditional interface as well as an interactive visual interface that does not support human-in-the-loop topic model.

There is increasing evidence that the effectiveness of information visualization techniques can be impacted by the particular needs and abilities of each user. This suggests that it is important to investigate information visualization systems that can dynamically adapt to each user. In this paper, we address the question of how to adapt. In particular, we present a study to evaluate a variety of visual prompts, called "interventions", that can be performed on a visualization to help users process it. Our results show that some of the tested interventions perform better than a condition in which no intervention is provided, both in terms of task performance as well as subjective user ratings. We also discuss findings on how intervention effectiveness is influenced by individual differences and task complexity.

Abstract There is increasing evidence that user characteristics can have a significant impact on visualization effectiveness, suggesting that visualizations could be designed to better fit each user's specific needs. Most studies to date, however, have looked at static visualizations. Studies considering interactive visualizations have only looked at a limited number of user characteristics, and consider either low‐level tasks (e.g., value retrieval), or high‐level tasks (in particular: discovery), but not both. This paper contributes to this line of work by looking at the impact of a large set of user characteristics on user performance with interactive visualizations, for both low and high‐level tasks. We focus on interactive visualizations that support decision making, exemplified by a visualization known as Value Charts. We include in the study two versions of ValueCharts that differ in terms of layout, to ascertain whether layout mediates the impact of individual differences and could be considered as a form of personalization. Our key findings are that (i) performance with low and high‐level tasks is affected by different user characteristics, and (ii) users with low visual working memory perform better with a horizontal layout. We discuss how these findings can inform the provision of personalized support to visualization processing.

Abstract Today it is quite common for people to exchange hundreds of comments in online conversations (e.g., blogs). Often, it can be very difficult to analyze and gain insights from such long conversations. To address this problem, we present a visual text analytic system that tightly integrates interactive visualization with novel text mining and summarization techniques to fulfill information needs of users in exploring conversations. At first, we perform a user requirement analysis for the domain of blog conversations to derive a set of design principles. Following these principles, we present an interface that visualizes a combination of various metadata and textual analysis results, supporting the user to interactively explore the blog conversations. We conclude with an informal user evaluation, which provides anecdotal evidence about the effectiveness of our system and directions for further design.