Chris North

Brian Keith Norambuena, Tanushree Mitra, Chris North

Narratives are fundamental to our understanding of the world, providing us with a natural structure for knowledge representation over time. Computational narrative extraction is a subfield of artificial intelligence that makes heavy use of information retrieval and natural language processing techniques. Despite the importance of computational narrative extraction, relatively little scholarly work exists on synthesizing previous research and strategizing future research in the area. In particular, this article focuses on extracting news narratives from an event-centric perspective. Extracting narratives from news data has multiple applications in understanding the evolving information landscape. This survey presents an extensive study of research in the area of event-based news narrative extraction. In particular, we screened over 900 articles that yielded 54 relevant articles. These articles are synthesized and organized by representation model, extraction criteria, and evaluation approaches. Based on the reviewed studies, we identify recent trends, open challenges, and potential research lines.

Brian Keith Norambuena, Tanushree Mitra, Chris North

Narrative sensemaking is an essential part of understanding sequential data. Narrative maps are a visual representation model that can assist analysts to understand narratives. In this work, we present a semantic interaction (SI) framework for narrative maps that can support analysts through their sensemaking process. In contrast to traditional SI systems which rely on dimensionality reduction and work on a projection space, our approach has an additional abstraction layer -- the structure space -- that builds upon the projection space and encodes the narrative in a discrete structure. This extra layer introduces additional challenges that must be addressed when integrating SI with the narrative extraction pipeline. We address these challenges by presenting the general concept of Mixed Multi-Model Semantic Interaction (3MSI) -- an SI pipeline, where the highest-level model corresponds to an abstract discrete structure and the lower-level models are continuous. To evaluate the performance of our 3MSI models for narrative maps, we present a quantitative simulation-based evaluation and a qualitative evaluation with case studies and expert feedback. We find that our SI system can model the analysts' intent and support incremental formalism for narrative maps.

Xuxin Tang, Ibrahim Tahmid, Eric Krokos et al.

Interactive spatial layouts empower users to synthesize information and organize findings for sensemaking. While Large Language Models (LLMs) can automate narrative generation from spatial layouts, current collage-based and re-generation methods struggle to support the incremental spatial refinements inherent to the sensemaking process. We identify three critical gaps in existing spatial-textual generation: interaction-revision misalignment, human-LLM intent misalignment, and lack of granular customization. To address these, we introduce Semantic Prompting, a framework for spatial refinement that perceives semantic interactions, reasons about refinement intent, and performs targeted positional revisions. We implemented S-PRISM to realize this framework. The empirical evaluation demonstrated that S-PRISM effectively enhanced the precision of interaction-revision refinement. A user study ($N=14$) highlighted how participants leveraged S-PRISM for incremental formalization through interactive steering. Results showed that users valued its efficient, adaptable, and trustworthy support, which effectively strengthens human-LLM intent alignment.

Daniel Palamarchuk, Lemara Williams, Brian Mayer et al.

Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documents can be directly analyzed in a temporal context. This paper proposes an expansion of the compass-aligned temporal Word2Vec methodology into dynamic topic modeling. Such a method allows for the direct comparison of word and document embeddings across time in dynamic topics. This enables the creation of visualizations that incorporate temporal word embeddings within the context of documents into topic visualizations. In experiments against the current state-of-the-art, our proposed method demonstrates overall competitive performance in topic relevancy and diversity across temporal datasets of varying size. Simultaneously, it provides insightful visualizations focused on temporal word embeddings while maintaining the insights provided by global topic evolution, advancing our understanding of how topics evolve over time.

Wei Liu, Eric Krokos, Kirsten Whitley et al.

Low-dimensional projections of text embeddings support visual analysis of document collections, but their spatial organization may not reflect the relationships an analyst intends to examine. Existing semantic interaction approaches encode semantic intent indirectly through geometric constraints or model updates, limiting interpretability and flexibility. We introduce LLM-augmented semantic steering, which enables analysts to express semantic intent by grouping a small set of example documents within the projection. A large language model externalizes this intent as natural-language representations and selectively extends it to related documents; the resulting semantic information is then incorporated into document representations via text augmentation or embedding-level blending, without retraining the underlying models. A case study illustrates how the same corpus can be reorganized from different semantic perspectives, while simulation-based evaluation shows that semantic steering improves global and local alignment with target semantic structures using only minimal interaction. Embedding-level blending further enables continuous and controllable steering of projection layouts. These results position projection spaces as intent-dependent semantic workspaces that can be reshaped through explicit, interpretable, language-mediated interaction.

Brian Keith, Fausto German, Eric Krokos et al.

As narrative extraction systems grow in complexity, establishing user trust through interpretable and explainable outputs becomes increasingly critical. This paper presents an evaluation of an Explainable Artificial Intelligence (XAI) system for narrative map extraction that provides meaningful explanations across multiple levels of abstraction. Our system integrates explanations based on topical clusters for low-level document relationships, connection explanations for event relationships, and high-level structure explanations for overall narrative patterns. In particular, we evaluate the XAI system through a user study involving 10 participants that examined narratives from the 2021 Cuban protests. The analysis of results demonstrates that participants using the explanations made the users trust in the system's decisions, with connection explanations and important event detection proving particularly effective at building user confidence. Survey responses indicate that the multi-level explanation approach helped users develop appropriate trust in the system's narrative extraction capabilities. This work advances the state-of-the-art in explainable narrative extraction while providing practical insights for developing reliable narrative extraction systems that support effective human-AI collaboration.

Brian Felipe Keith-Norambuena, Fausto German, Eric Krokos et al.

Semantic interaction (SI) enables analysts to incorporate their cognitive processes into AI models through direct manipulation of visualizations. While SI frameworks for narrative extraction have been proposed, empirical evaluations of their effectiveness remain limited. This paper presents a user study that evaluates SI for narrative map sensemaking, involving 33 participants under three conditions: a timeline baseline, a basic narrative map, and an interactive narrative map with SI capabilities. The results show that the map-based prototypes yielded more insights than the timeline baseline, with the SI-enabled condition reaching statistical significance and the basic map condition trending in the same direction. The SI-enabled condition showed the highest mean performance; differences between the map conditions were not statistically significant but showed large effect sizes (d > 0.8), suggesting that the study was underpowered to detect them. Qualitative analysis identified two distinct SI approaches-corrective and additive-that enable analysts to impose quality judgments and organizational structure on extracted narratives. We also find that SI users achieved comparable exploration breadth with less parameter manipulation, suggesting that SI serves as an alternative pathway for model refinement. This work provides empirical evidence that map-based representations outperform timelines for narrative sensemaking, along with qualitative insights into how analysts use SI for narrative refinement.

Yali Bian, Chris North

In this paper, we design novel interactive deep learning methods to improve semantic interactions in visual analytics applications. The ability of semantic interaction to infer analysts' precise intents during sensemaking is dependent on the quality of the underlying data representation. We propose the $\text{DeepSI}_{\text{finetune}}$ framework that integrates deep learning into the human-in-the-loop interactive sensemaking pipeline, with two important properties. First, deep learning extracts meaningful representations from raw data, which improves semantic interaction inference. Second, semantic interactions are exploited to fine-tune the deep learning representations, which then further improves semantic interaction inference. This feedback loop between human interaction and deep learning enables efficient learning of user- and task-specific representations. To evaluate the advantage of embedding the deep learning within the semantic interaction loop, we compare $\text{DeepSI}_{\text{finetune}}$ against a state-of-the-art but more basic use of deep learning as only a feature extractor pre-processed outside of the interactive loop. Results of two complementary studies, a human-centered qualitative case study and an algorithm-centered simulation-based quantitative experiment, show that $\text{DeepSI}_{\text{finetune}}$ more accurately captures users' complex mental models with fewer interactions.

Brian Felipe Keith Norambuena, Tanushree Mitra, Chris North

Narrative sensemaking is a fundamental process to understand sequential information. Narrative maps are a visual representation framework that can aid analysts in their narrative sensemaking process. Narrative maps allow analysts to understand the big picture of a narrative, uncover new relationships between events, and model the connection between storylines. We seek to understand how analysts create and use narrative maps in order to obtain design guidelines for an interactive visualization tool for narrative maps that can aid analysts in narrative sensemaking. We perform two experiments with a data set of news articles. The insights extracted from our studies can be used to design narrative maps, extraction algorithms, and visual analytics tools to support the narrative sensemaking process. The contributions of this paper are three-fold: (1) an analysis of how analysts construct narrative maps; (2) a user evaluation of specific narrative map features; and (3) design guidelines for narrative maps. Our findings suggest ways for designing narrative maps and extraction algorithms, as well as providing insights towards useful interactions. We discuss these insights and design guidelines and reflect on the potential challenges involved. As key highlights, we find that narrative maps should avoid redundant connections that can be inferred by using the transitive property of event connections, reducing the overall complexity of the map. Moreover, narrative maps should use multiple types of cognitive connections between events such as topical and causal connections, as this emulates the strategies that analysts use in the narrative sensemaking process.

Brian Felipe Keith Norambuena, Tanushree Mitra, Chris North

Narrative sensemaking is a fundamental process to understand sequential information. Narrative maps are a visual representation framework that can aid analysts in this process. They allow analysts to understand the big picture of a narrative, uncover new relationships between events, and model connections between storylines. As a sensemaking tool, narrative maps have applications in intelligence analysis, misinformation modeling, and computational journalism. In this work, we seek to understand how analysts construct narrative maps in order to improve narrative map representation and extraction methods. We perform an experiment with a data set of news articles. Our main contribution is an analysis of how analysts construct narrative maps. The insights extracted from our study can be used to design narrative map visualizations, extraction algorithms, and visual analytics tools to support the sensemaking process.

Sorour E. Amiri, Bijaya Adhikari, John Wenskovitch et al.

Generating useful network summaries is a challenging and important problem with several applications like sensemaking, visualization, and compression. However, most of the current work in this space do not take human feedback into account while generating summaries. Consider an intelligence analysis scenario, where the analyst is exploring a similarity network between documents. The analyst can express her agreement/disagreement with the visualization of the network summary via iterative feedback, e.g. closing or moving documents ("nodes") together. How can we use this feedback to improve the network summary quality? In this paper, we present NetReAct, a novel interactive network summarization algorithm which supports the visualization of networks induced by text corpora to perform sensemaking. NetReAct incorporates human feedback with reinforcement learning to summarize and visualize document networks. Using scenarios from two datasets, we show how NetReAct is successful in generating high-quality summaries and visualizations that reveal hidden patterns better than other non-trivial baselines.

John Wenskovitch, Chris North

How do analysts think about grouping and spatial operations? This overarching question incorporates a number of points for investigation, including understanding how analysts begin to explore a dataset, the types of grouping/spatial structures created and the operations performed on them, the relationship between grouping and spatial structures, the decisions analysts make when exploring individual observations, and the role of external information. This work contributes the design and results of such a study, in which a group of participants are asked to organize the data contained within an unfamiliar quantitative dataset. We identify several overarching approaches taken by participants to design their organizational space, discuss the interactions performed by the participants, and propose design recommendations to improve the usability of future high-dimensional data exploration tools that make use of grouping (clustering) and spatial (dimension reduction) operations.

Yali Bian, Michelle Dowling, Chris North

Semantic interaction (SI) attempts to learn the user's cognitive intents as they directly manipulate data projections during sensemaking activity. For text analysis, prior implementations of SI have used common data features, such as bag-of-words representations, for machine learning from user interactions. Instead, we hypothesize that features derived from deep learning word embeddings will enable SI to better capture the user's subtle intents. However, evaluating these effects is difficult. SI systems are usually evaluated by a human-centred qualitative approach, by observing the utility and effectiveness of the application for end-users. This approach has drawbacks in terms of replicability, scalability, and objectiveness, which makes it hard to perform convincing contrast experiments between different SI models. To tackle this problem, we explore a quantitative algorithm-centered analysis as a complementary evaluation approach, by simulating users' interactions and calculating the accuracy of the learned model. We use these methods to compare word-embeddings to bag-of-words features for SI.

Yali Bian, John Wenskovitch, Chris North

This paper examines how deep learning (DL) representations, in contrast to traditional engineered features, can support semantic interaction (SI) in visual analytics. SI attempts to model user's cognitive reasoning via their interaction with data items, based on the data features. We hypothesize that DL representations contain meaningful high-level abstractions that can better capture users' high-level cognitive intent. To bridge the gap between cognition and computation in visual analytics, we propose DeepVA (Deep Visual Analytics), which uses high-level deep learning representations for semantic interaction instead of low-level hand-crafted data features. To evaluate DeepVA and compare to SI models with lower-level features, we design and implement a system that extends a traditional SI pipeline with features at three different levels of abstraction. To test the relationship between task abstraction and feature abstraction in SI, we perform visual concept learning tasks at three different task abstraction levels, using semantic interaction with three different feature abstraction levels. DeepVA effectively hastened interactive convergence between cognitive understanding and computational modeling of the data, especially in high abstraction tasks.

Andrew McCaleb Reach, Chris North

Animation is ubiquitous in visualization systems, and a common technique for creating these animations is the transition. In the transition approach, animations are created by smoothly interpolating a visual attribute between a start and end value, reaching the end value after a specified duration. This approach works well when each transition for an attribute is allowed to finish before the next is triggered, but performs poorly when a new transition is triggered before the current transition has finished. In particular, interruptions introduce velocity discontinuities, and frequent interruptions can slow down the resulting animation. To solve these problems, we model the problem of animation as a signal processing problem. In our technique, animations are produced by transformations of signals, or functions over time. In particular, an animation is produced by transforming an input signal, a function from time to target attribute value, into an output signal, a function from time to displayed attribute value. We show that well-known signal-processing techniques can be applied to produce animations that are free from velocity discontinuities even when interrupted.

Hao Wu, Maoyuan Sun, Peng Mi et al.

Modern visual analytic tools promote human-in-the-loop analysis but are limited in their ability to direct the user toward interesting and promising directions of study. This problem is especially acute when the analysis task is exploratory in nature, e.g., the discovery of potentially coordinated relationships in massive text datasets. Such tasks are very common in domains like intelligence analysis and security forensics where the goal is to uncover surprising coalitions bridging multiple types of relations. We introduce new maximum entropy models to discover surprising chains of relationships leveraging count data about entity occurrences in documents. These models are embedded in a visual analytic system called MERCER that treats relationship bundles as first class objects and directs the user toward promising lines of inquiry. We demonstrate how user input can judiciously direct analysis toward valid conclusions whereas a purely algorithmic approach could be led astray. Experimental results on both synthetic and real datasets from the intelligence community are presented.