Mining and conveying actionable insights from complex data is a key challenge of exploratory data analysis (EDA) and storytelling.To address this challenge, we present a design space for actionable EDA and storytelling.Synthesizing theory and expert interviews, we highlight how semantic precision, rhetorical persuasion, and pragmatic relevance underpin effective EDA and storytelling.We also show how this design space subsumes common challenges in actionable EDA and storytelling, such as identifying appropriate analytical strategies and leveraging relevant domain knowledge.Building on the potential of LLMs to generate coherent narratives with commonsense reasoning, we contribute Jupybara, an AI-enabled assistant for actionable EDA and storytelling implemented as a Jupyter Notebook extension.Jupybara employs two strategiesdesign-space-aware prompting and multi-agent architectures-to operationalize our design space.An expert evaluation confirms Jupybara's usability, steerability, explainability, and reparability, as well as the effectiveness of our strategies in operationalizing the design space framework with LLMs.

Conversational Recommendation Systems (CRSs) are a particularly interesting application for out-of-the-box LLMs due to their potential for eliciting user preferences and making recommendations in natural language across a wide set of domains. Somewhat surprisingly, we find however that in such a conversational application, the more questions a user answers about their preferences, the worse the model’s recommendations become. We demonstrate this phenomenon on a previously published dataset as well as two novel datasets which we contribute. We also explain why earlier benchmarks failed to detect this round-over-round performance loss, highlighting the importance of the evaluation strategy we use and expanding upon Li et al. (2023a). We also present preference elicitation and recommendation strategies that mitigate this degradation in performance, beating state-of-the-art results, and show how three underlying models, GPT-3.5, GPT-4, and Claude 3.5 Sonnet, differently impact these strategies. Our datasets and code are available at https://github.com/CtrlVGustavo/A-Flash- in-the-Pan-CRS.

We consider behavior of agents in a long-term multiagent coopetitive setting in which agents vary their cooperative and competitive stances over time. Using the game ofDiplomacyas a testbed, we study how successful agents vary their coopetitive behavior, developing a new “style of play” (SoP) characterization of player behavior. We assess five novel SoP hypotheses about successful behavior. We propose two algorithms to automatically compute an agent’s SoP vector and describe the important factors in this computation. As an agent’s SoP depends on the game state and its perception of threat, we develop a novel “means, motive, and opportunity” (MMO) model of threat and show that we can predict threats effectively using this model. We provide novel insights into how agents should behave to more successfully achieve their goals in long-term coopetitive settings.

Conversations aimed at determining good recommendations are iterative in nature.People often express their preferences in terms of a critique of the current recommendation (e.g., "It doesn't look good for a date"), requiring some degree of common sense for a preference to be inferred.In this work, we present a method for transforming a user critique into a positive preference (e.g., "I prefer more romantic") in order to retrieve reviews pertaining to potentially better recommendations (e.g., "Perfect for a romantic dinner").We leverage a large neural language model (LM) in a fewshot setting to perform critique-to-preference transformation, and we test two methods for retrieving recommendations: one that matches embeddings, and another that fine-tunes an LM for the task.We instantiate this approach in the restaurant domain and evaluate it using a new dataset of restaurant critiques.In an ablation study, we show that utilizing critiqueto-preference transformation improves recommendations, and that there are at least three general cases that explain this improved performance.

There is growing concern about the use of social platforms to push political narratives during elections. One very recent case is Brazil's, where WhatsApp is now widely perceived as a key enabler of the far-right's rise to power. In this paper, we perform a large-scale analysis of partisan WhatsApp groups to shed light on how both right-wingers and left-wingers used the platform in the 2018 Brazilian presidential election. Across its two rounds, we collected +2.8M messages from +45k users in 232 public groups (175 right-wing vs. 57 left-wing). After describing how we obtained a sample that is many times larger than previous works, we contrast right-wingers and left-wingers on their social network metrics, regional distribution of users, content-sharing habits, and most characteristic news sources.

News articles often use narrative frames to present people, organizations, and facts. These narrative frames follow cultural archetypes, enabling readers to associate each of the presented elements with familiar stereotypes, well-known characters, and recognizable outcomes. In this way, authors can cast real people or organizations as heroes, villains, or victims. We present a system that identifies the main entities of a news article, and determines which is being cast as a hero, a villain, or a victim. As currently implemented, this system interacts directly with news consumers through a browser extension. Our hope is that by informing readers when an entity is cast in one of these roles, we can make implicit bias explicit, and thereby assist readers in applying their media literacy skills. This approach can also be used to identify roles in well-understood event sequences in a more prosaic manner, e.g., for information extraction.

Distributed representations of words have been shown to capture lexical semantics, based on their effectiveness in word similarity and analogical relation tasks. But, these tasks only evaluate lexical semantics indirectly. In this paper, we study whether it is possible to utilize distributed representations to generate dictionary definitions of words, as a more direct and transparent representation of the embeddings' semantics. We introduce definition modeling, the task of generating a definition for a given word and its embedding. We present different definition model architectures based on recurrent neural networks, and experiment with the models over multiple data sets. Our results show that a model that controls dependencies between the word being defined and the definition words performs significantly better, and that a character-level convolution layer that leverages morphology can complement word-level embeddings. Our analysis reveals which components of our models contribute to accuracy. Finally, the errors made by a definition model may provide insight into the shortcomings of word embeddings.

Whether it's a social media system populating a news feed or a user searching for content to share, decisions are constantly being made on the basis of semantic information. Topic, sentiment, and preference are among the many semantic dimensions of content that humans and machines must carefully weigh when prioritizing media consumption. Though easy for humans, such natural language tasks are nontrivial computationally. In this paper, we present a novel technique for sorting a corpus of news articles based on two competing semantic objectives. Solving this multi-objective optimization problem yields a pareto front with a finite set of solution articles. Iterating on the remaining data, we construct solution sets in tiers of successive pareto fronts. Our technique allows for exploration of the tradeoffs between semantic concepts, yielding cogent and trope-like results.

AttitudeBuzz is a system that analyzes and presents complex social attitudes based on geolocated social media data. The system uses a machine learning model to apply highly domain-specific sentiment analysis to such data, specifically Twitter, by learning modulators around a configurable lexicon central to the domain of inquiry. Training data are acquired from geographical areas where a specific attitude or opinion is known to dominate. We apply AttitudeBuzz to the domain of homophobic attitudes expressed on Twitter. The resulting user interface is presented and the machine learning model described and analyzed.

Probably the major approach to making predictions or recommendations about user behavior is by pairing unambiguous indicators of preference with attributes of the users who have given those indications. However, since the necessary preference data is often difficult to obtain, it is considered very valuable and often held closely by vendors and advertisers. On the other hand, while Twitter and other social media platforms provide a wealth of data about users by way of what they say or tweet, who or what they like or follow, etc., little work has been done to combine these data with indicators of preference for purposes of prediction or recommendation. In this paper we present a novel approach to mining preference data from natural language expressions in social media, which are then extrapolated to other individuals whose preferences are not known through predictive modeling. As an application for this approach, we describe the implementation of Tweetcast Your Vote, a publicly accessible system that predicted the voting decisions of Twitter users in the 2012 U.S. presidential election.

The Web makes it possible for news readers to learn more about virtually any story that interests them. Media outlets and search engines typically augment their information with links to similar stories. It is up to the user to determine what new information is added by them, if any. In this paper we present Tell Me More, a system that performs this task automatically: given a seed news story, it mines the web for similar stories reported by different sources and selects snippets of text from those stories which offer new information beyond the seed story. New content may be classified as supplying: additional quotes, additional actors, additional figures and additional information depending on the criteria used to select it. In this paper we describe how the system identifies new and informative content with respect to a news story. We also how that providing an explicit categorization of new information is more useful than a binary classification (new/not-new). Lastly, we show encouraging results from a preliminary evaluation of the system that validates our approach and encourages further study.

Using content-specific models to guide information retrieval and extraction can provide richer interfaces to end-users for both understanding the context of news events and navigating related news articles. In this paper we discuss a system, Brussell, that uses semantic models to organize retrieval and extraction results, generating both storylines explaining how news event situations unfold and also biographical sketches of the situation participants. We generalize these models to introduce a new category of knowledge representation, an explanatory structure, that can scale up to include information from hundreds of documents, yet still provide model-based UI support to end-users. An informal survey of business news suggests the broad prevalence of news event situations indicating Brussell's potential utility, while an evaluation quantifies its performance in finding kidnapping situations.

Readers interested in the context of an event covered in the news such as the dismissal of a lawsuit can benefit from easily finding out about the overall news situation, the legal trial, of which the event is a part. Guided by abstract models of news situation types such as legal trials, corporate acquisitions, and kidnappings, Brussell is a system that presents situation instances it creates by reading multiple articles about the specific events that comprise them. We discuss how these situation models are structured and how they drive the creation of particular instances.

Blogs have become an important medium for people to express opinions and share information on the web. Predicting the interests of bloggers can be beneficial for information retrieval and knowledge discovery in the blogosphere. In this paper, we propose a two-layer classification model to categorize the interests of bloggers based on a set of short snippets collected from their blog posts. Experiments were conducted on a list of bloggers collected from blog directories, with their snippets collected from Google Blog Search. The results show that the proposed method is robust to errors in the lower level and achieve satisfactory performance in categorizing blogger's interests.

The web has become an important medium for news delivery and consumption. Fresh content about a variety of topics and events is constantly being created and published on the web by many sources. As intuitively understood by readers, and studied in journalism, news articles produced by different social groups present different attitudes towards and interpretations of the same news issues. In this paper, we propose a new paradigm for aggregating news articles according to the news sources related to the stakeholders of the news issues. We implement this paradigm in a prototype system called LocalSavvy. The system provides users the capability to aggregate and browse various local views about the news issues in which they are interested.

We propose a machine learning system that learns to choose human gestures to accompany novel text. The system is trained on scripts comprised of speech and animations that were hand-coded by professional animators and shipped in video games. We treat this as a text-classification problem, classifying speech as corresponding with specific classes of gestures. We have built and tested two separate classifiers. The first is trained simply on the frequencies of different animations in the corpus. The second extracts text features from each script, and maps these features to the gestures that accompany the script. We have experimented with using a number of features of the text, including n-grams, emotional valence of the text, and parts-of-speech. Using a naïve Bayes classifier, the system learns to associate these features with appropriate classes of gestures. Once trained, the system can be given novel text for which it will attempt to assign appropriate gestures. We examine the performance of the two classifiers by using n-fold cross-validation over our training data, as well as two user studies of subjective evaluation of the results. Although there are many possible applications of automated gesture assignment, we hope to apply this technique to a system that produces an automated news show.

The importance of named entities in information retrieval and knowledge management has recently brought interest in characterizing semantic relationships between entities. In this paper, we propose a method for measuring semantic similarity, an important type of semantic relationship, between entities. The method is based on Google Directory, a search interface to the Open Directory Project. Via the search engine, we can locate the web pages relevant to an entity and automatically create a profile of the entity according to the directory assignments of its web pages, which capture various features of the entity. Using their profiles, the semantic similarity between entities can be measured in different dimensions. We apply the semantic similarity measurement to two knowledge acquisition tasks: thesaurus construction of entities and fine grained categorization of entities. Our experiments demonstrate that the proposed method works effectively in these two tasks.

Comparing and contrasting is an important strategy people employ to understand new situations and create solutions for new problems. Similar events can provide hints for problem solving, as well as larger contexts for understanding the specific circumstances of an event. Lessons can leaned from past experience, insights can be gained about the new situation from familiar examples, and trends can be discovered among similar events. As the largest knowledge base for human beings, the Web provides both an opportunity and a challenge to discover comparable cases in order to facilitate situation analysis and problem solving. In this paper, we present Compare & Contrast, a system that uses the Web to discover comparable cases for news stories, documents about similar situations but involving distinct entities. The system analyzes a news story given by the user and builds a model of the story. With the story model, the system dynamically discovers entities comparable to the main entity in the original story and uses these comparable entities as seeds to retrieve web pages about comparable cases. The system is domain independent, does not require any domain-specific knowledge engineering efforts, and deals with the complexity of unstructured text and noise on the web in a robust way. We evaluated the system with an experiment on a collection of news articles and a user study.

Pinpoint is a promising first step towards using a rich model of task context in proactive and dynamic IR systems. Pinpoint allows a user to navigate decision tree representations of problem spaces, built by domain experts, while dynamically entering annotations specific to their problem. The system then automatically generates queries to information repositories based on both the user's annotations and location in the problem space, producing results that are both task focused and problem specific. Initial feedback from users and domain experts has been positive.

In this article, we describe the intrinsic constraints of mobility and discuss how we can work around and often exploit these constraints using information implied by the context of the mobile user. In particular, we outline some of the work we have been doing in providing information to users on the basis of "Low-Fi" location information retrieved from user's cellular phone. A user's location and other contextual information is used to retrieve task relevant information and avoid many of the problems posed by limits of mobile devices and their use. This is part of an ongoing effort to build "smart" navigational interfaces bound on information about a user's location.

The work presented in this paper takes a novel approach to the task of providing information to viewers of broadcast news. Instead of considering the broadcast news as the end product, this work uses it as a starting point to dynamically build an information space for the user to explore. This information space is designed to satisfy the users information needs, by containing more breadth, depth, and points of view than the original broadcast story. The architecture and current implementation are discussed, and preliminary results from the analysis of some its components are presented

This research discusses a method for delivering just-in-time information to television viewers to provide more depth and more breadth to television broadcasts. A novel aspect of this research is that it uses broadcast news as a starting point for gathering information regarding specific stories, as opposed to considering the broadcast version to be the end of the viewers exploration. This work is implemented in Cronkite, a system that provides viewers with expanded coverage of broadcast news stories.

In this paper, we describe the Interactive Chef (IChef), an intelligent system that attempts to emulate a human cooking assistant. The user navigates through a recipe using voice, while IChef reads aloud each step. IChef is designed to be aware of the present context and exploit this knowledge by answering questions about ingredients or techniques posed by the user in natural spoken language.

mpME! introduces users to new music and interweaves old music the listener has heard of before and enjoys. mpME! exposes listeners to new music by using existing music information on the Internet such as musical artist directories. mpME! also provides complementary information about the artist to educate the user throughout the listening experience. MpME! uses a feature intersection algorithm to make recommendations for artists using features of artists found on the web.

Contents of documents can be characterized not just by subject matter or topic, but also in terms of more thematic or even stylistic attributes, such as level of detail, emotional vs. unemotional presentation, personal vs. impersonal voice, positive or negative attitude or "spin," theoretical vs. practical presentation, etc. As implied by these examples, such attributes often form continuous dimensions along which documents can be distinguished. In the best case, the appropriate place on these dimensions -- i.e., which documents will be of most value to the user at the current time -- can be determined by the user's current task, prior knowledge, etc. Does the user need more detail, or less, for example? In the current state of the art, it is difficult to compute this automatically. Instead, we propose that users should be enabled view these dimensions using themometers and to navigate them using themostats -- controls that offer users a choice of documents on the same topic, but with differing levels of detail, positive or negative spin, personal or impersonal voice, etc. The dimensions along which we can provide this kind of control must be easily computable. We have been able to characterize documents with respect to a number of these dimensions using simple statistical measures as well as specialized dictionaries, and describe these methods herein.

While recent work has focused on providing tools and infrastructure for users to access electronic information over the Internet, the relationship between the physical world and information available online has been relatively unexplored. Information about a user's location, and the objects she interacts with, can be sufficient to recognize enough of the user's task to drive retrieval of online information relevant to the task at hand. The XLibris system automatically retrieves, aggregates, and delivers information about books to users as they are checked out of the library, using information about the books themselves and the user's task. XLibris locates books in the Dewey Decimal subject hierarchy to automatically search for the most relevant information about the book for the user, tailoring both the sources queried and the information returned based on the book's position in the hierarchy.

Many case-based reasoning systems have been developed to aid designers by providing them with libraries of prior design experiences. Traditionally, these systems are implemented as stand-alone ``external memories'' for the designer to query manually. This paper presents a contrasting approach that integrates proactive case retrieval into the designer's normal task processes, automatically tailoring information selection and presentation emphasis to fit changing designer needs and attention flow. The paper presents a set of principles for this integrated intelligent design support and describes their application in the Stamping Advisor, a system to support design feasibility analysis for sheet metal automotive parts. The Stamping Advisor interface proactively provides designers with relevant information to support feasibility analysis, automatically prepares their information products, and unobtrusively gathers the information needed to generate new cases to improve the quality of future support.

Article Integrating organizational memory and performance support Share on Authors: Christopher Johnson Northwestern University, The Institute for the Learning Sciences, 1890 Maple Avenue, Evanston, IL Northwestern University, The Institute for the Learning Sciences, 1890 Maple Avenue, Evanston, ILView Profile , Larry Birnbaum Northwestern University, The Institute for the Learning Sciences, 1890 Maple Avenue, Evanston, IL Northwestern University, The Institute for the Learning Sciences, 1890 Maple Avenue, Evanston, ILView Profile , Ray Bareiss Cognitive Arts Corporation, 120 South Riverside Plaza, Suite 1520, Chicago, IL Cognitive Arts Corporation, 120 South Riverside Plaza, Suite 1520, Chicago, ILView Profile , Tom Hinrichs Cognitive Arts Corporation, 120 South Riverside Plaza, Suite 1520, Chicago, IL Cognitive Arts Corporation, 120 South Riverside Plaza, Suite 1520, Chicago, ILView Profile Authors Info & Claims IUI '99: Proceedings of the 4th international conference on Intelligent user interfacesDecember 1998 Pages 127–134https://doi.org/10.1145/291080.291102Published:01 December 1998 5citation393DownloadsMetricsTotal Citations5Total Downloads393Last 12 Months3Last 6 weeks1 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

No abstract available.

Article Free Access Share on Interface design based on standardized task models Authors: Larry Birnbaum Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, IL Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, ILView Profile , Ray Bareiss Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, IL Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, ILView Profile , Tom Hinrichs Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, IL Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, ILView Profile , Christopher Johnson Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, IL Northwestern University, The Institute for the Learning Sciences, 1890 Maple Ave., Evanston, ILView Profile Authors Info & Claims IUI '98: Proceedings of the 3rd international conference on Intelligent user interfacesJanuary 1998 Pages 65–72https://doi.org/10.1145/268389.268400Published:01 January 1998Publication History 7citation650DownloadsMetricsTotal Citations7Total Downloads650Last 12 Months6Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

Article Compelling intelligent user interfaces—how much AI? Share on Authors: Larry Birnbaum ILS/Northwestern, 1890 Maple Avenue, Evanston IL ILS/Northwestern, 1890 Maple Avenue, Evanston ILView Profile , Eric Horvitz Microsoft Research, One Microsoft Way, Redmond WA Microsoft Research, One Microsoft Way, Redmond WAView Profile , David Kurlander Microsoft Research, One Microsoft Way, Redmond WA Microsoft Research, One Microsoft Way, Redmond WAView Profile , Henry Lieberman MIT Media Laboratory, 20 Ames Street, Cambridge MA MIT Media Laboratory, 20 Ames Street, Cambridge MAView Profile , Joe Marks MERL, 201 Broadway, Cambridge MA MERL, 201 Broadway, Cambridge MAView Profile , Steve Roth Robotics Institute, Carnegie Mellon University, Pittsburgh PA Robotics Institute, Carnegie Mellon University, Pittsburgh PAView Profile Authors Info & Claims IUI '97: Proceedings of the 2nd international conference on Intelligent user interfacesJanuary 1997 Pages 173–175https://doi.org/10.1145/238218.238319Online:06 January 1997Publication History 16citation637DownloadsMetricsTotal Citations16Total Downloads637Last 12 Months41Last 6 weeks3 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Vision should provide an explanation of the scene in terms of a causal semantics. The authors propose a characterization of what constitutes visual understanding. The cornerstone of the proposal is that visual understanding is fundamentally a matter of developing a causal explanation of the scene, i.e., of determining the causal significance of the elements in a scene, and the causal relationships among those elements. Simple, naive physical knowledge is used as the basis of a vertically integrated vision system that explains arbitrarily complex stacked block structures. The semantics provides a basis for controlling the application of visual attention, and forms a framework for the explanation that is generated. It is shown that the program sequentially explores scenes of complex blocks structures, identifies functional substructures such as arches and cantilevers, and develops an explanation of why the whole construction stands and the role of each block in its stability.>