The Text-to-SQL problem aims at developing natural language query interfaces for relational database systems by converting the text input into executable SQL queries. Recently, using Large Language Models (LLM) has emerged as a new paradigm for the Text-to-SQL problem. To this end, the LLM needs to understand not only user input but also information from the database. In this demo, we present multi-agent SQL (MageSQL), an LLM based Text-to-SQL approach that tackles the task by orchestrating multiple agents in a pipeline. We will showcase a user-friendly interface to demonstrate the inner workings of our approach that allows users to add and modify the agents with different functionalities, customize prompts, and see their impact on specific examples. Through several use cases, we will demonstrate how to (i) construct a Text-to-SQL pipeline with multiple agents; (ii) generate prompts for LLM with various templates and strategies; and (iii) monitor the results of natural language queries and perform debugging.

Information extraction (IE) approaches often play a pivotal role in text analysis and require significant human intervention. Therefore, a deeper understanding of existing IE practices and related challenges from a human-in-the-loop perspective is warranted. In this work, we conducted semi-structured interviews in an industrial environment and analyzed the reported IE approaches and limitations. We observed that data science workers often follow an iterative task model consisting of information foraging and sensemaking loops across all the phases of an IE workflow. The task model is generalizable and captures diverse goals across these phases (e.g., data preparation, modeling, evaluation.) We found several limitations in both foraging (e.g., data exploration) and sensemaking (e.g., qualitative debugging) loops stemming from a lack of adherence to existing cognitive engineering principles. Moreover, we identified that due to the iterative nature of an IE workflow, the requirement of provenance is often implied but rarely supported by existing systems. Based on these findings, we discuss design implications for supporting IE workflows and future research directions.

The rapid growth in biological sequence data is revolutionizing our understanding of genotypic diversity and challenging conventional approaches to informatics. With the increasing availability of genomic data, traditional bioinformatic tools require substantial computational time and the creation of ever-larger indices each time a researcher seeks to gain insight from the data. To address these challenges, we pre-computed important relationships between biological entities spanning the Central Dogma of Molecular Biology and captured this information in a relational database. The database can be queried across hundreds of millions of entities and returns results in a fraction of the time required by traditional methods. In this paper, we describe Functional Genomics Platform (formerly known as OMXWare), a comprehensive database relating genotype to phenotype for bacterial life. Continually updated, the Functional Genomics Platform today contains data derived from 200,000 curated, self-consistently assembled genomes. The database stores functional data for over 68 million genes, 52 million proteins, and 239 million domains with associated biological activity annotations from Gene Ontology, KEGG, MetaCyc, and Reactome. The Functional Genomics Platform maps all of the many-to-many connections between each biological entity including the originating genome, gene, protein, and protein domain. Various microbial studies, from infectious disease to environmental health, can benefit from the rich data and connections. We describe the data selection, the pipeline to create and update the Functional Genomics Platform, and the developer tools (Python SDK and REST APIs)which allow researchers to efficiently study microbial life at scale.

Natural-language interfaces are gaining popularity due to their potential to democratize access to data and insights by making the interaction with data more natural and accessible for a wide range of business users. To fully embrace the goal of democratization, it is also necessary to provide effective and continuous guidance support for data exploration. Conversational interfaces enable exploration of the data and insights search space in small incremental steps as the conversation with the data progresses. In this demo, we describe Boomerang, a system that recommends data-driven insights to guide exploration of datasets through a conversational interface. Boomerang aggregates recommendations from a variety of statistical, collaborative, and content-based recommenders, and selects insights that match closely to the user's current state of data exploration, represented as the \em conversational context. Boomerang combines various metrics, such as \em relevance, \em interestingness and \em timeliness, to rank the insights and recommends the insights based on current conversational context. In the demo, we will show how Boomerang enables guided data exploration on a sales dataset, containing information about products, retailers, sales, orders, inventory levels and regions.

Open data analysis platforms are being adopted to support collaboration in science and business. Studies suggest that analytic work in an enterprise occurs in a complex ecosystem of people, data, and software working in a coordinated manner. These studies also point to friction between the elements of this ecosystem that reduces user productivity and quality of work. LabBook is an open, social, and collaborative data analysis platform designed explicitly to reduce this friction and accelerate discovery. Its goal is to help users leverage each other's knowledge and experience to find the data, tools and collaborators they need to integrate, visualize, and analyze data. The key insight is to collect and use more metadata about all elements of the analytic ecosystem by means of an architecture and user experience that reduce the cost of contributing such metadata. We demonstrate how metadata can be exploited to improve the collaborative user experience and facilitate collaborative data integration and recommendations. We describe a specific use case and discuss several design issues concerning the capture, representation, querying and use of metadata.

We developed DataBoard, a freeform spatial interface, to support users in simple problem solving tasks. To develop a deeper understanding of the role of space and the tradeoffs between freeform and structured interaction styles in problem solving tasks, we conducted a controlled user study comparing the DataBoard with a spreadsheet and analyzed video data in detail. Beyond improvements in task performance and memory recall, our observations reveal that freeform interfaces can support users in a variety of ways: representing problems flexibly, developing strategies, executing strategies incrementally, tracking problem state easily, reducing mental computation, and verifying solutions perceptually. The spreadsheet also had advantages, and we discuss the tradeoffs.

We present Koala, a system that enables users to capture, share, automate, and personalize business processes on the web. Koala is a collaborative programming-by-demonstration system that records, edits, and plays back user interactions as pseudo-natural language scripts that are both human- and machine-interpretable. Unlike previous programming by demonstration systems, Koala leverages sloppy programming that interprets pseudo-natural language instructions (as opposed to formal syntactic statements) in the context of a given web page's elements and actions. Koala scripts are automatically stored in the Koalescence wiki, where a community of users can share, run, and collaboratively develop their "how-to" knowledge. Koala also takes advantage of corporate and personal data stores to automatically generalize and instantiate user-specific data, so that scripts created by one user are automatically personalized for others. Our initial experiences suggest that Koala is surprisingly effective at interpreting instructions originally written for people.

We present Avatar Semantic Search, a prototype search engine that exploits annotations in the context of classical keyword search. The process of annotations is accomplished offline by using high-precision information extraction techniques to extract facts, con-cepts, and relationships from text. These facts and concepts are represented and indexed in a structured data store. At runtime, keyword queries are interpreted in the context of these extracted facts and converted into one or more precise queries over the structured store. In this demonstration we describe the overall architecture of the Avatar Semantic Search engine. We also demonstrate the superiority of the AVATAR approach over traditional keyword search engines using Enron email data set and a blog corpus.

System administrators work with many different tools to manage and fix complex hardware and software infrastructure in a rapidly paced work environment. Through extensive field studies, we observed that they often build and share custom tools for specific tasks that are not supported by vendor tools. Recent trends toward web-based management consoles offer many advantages but put an extra burden on system administrators, as customization requires web programming, which is beyond the skills of many system administrators. To meet their needs, we developed A1, a spreadsheet-based environment with a task-specific system-administration language for quickly creating small tools or migrating existing scripts to run as web portlets. Using A1, system administrators can build spreadsheets to access remote and heterogeneous systems, gather and integrate status data, and orchestrate control of disparate systems in a uniform way. A preliminary user study showed that in just a few hours, system administrators can learn to use A1 to build relatively complex tools from scratch.

Computer system administrators are the unsung heroes of the information age, working behind the scenes to configure, maintain, and troubleshoot the computer infrastructure that underlies much of modern life. However, little can be found in the literature about the practices and problems of these highly specialized computer users. We conducted a series of field studies in large corporate data centers, observing organizations, work practices, tools, and problem-solving strategies of system administrators. We found system administrators operate within large-scale, complex environments that present significant technical, social, cognitive, and business challenges. In this paper, we describe system administrator tool use in critical, high-cost, labor-intensive work through observational, survey, and interview data. We discuss our findings concerning administrator needs for coordinating work, maintaining situation awareness, planning and rehearsing complex procedures, building tools, and supporting complicated interleaved workflows.

Interactive visualizations are effective tools in mining scientific, engineering, and business data to support decision-making activities. Star Coordinates is proposed as a new multi-dimensional visualization technique, which supports various interactions to stimulate visual thinking in early stages of knowledge discovery process. In Star Coordinates, coordinate axes are arranged on a two-dimensional surface, where each axis shares the same origin point. Each multi-dimensional data element is represented by a point, where each attribute of the data contributes to its location through uniform encoding. Interaction features of Star Coordinates provide users the ability to apply various transformations dynamically, integrate and separate dimensions, analyze correlations of multiple dimensions, view clusters, trends, and outliers in the distribution of data, and query points based on data ranges. Our experience with Star Coordinates shows that it is particularly useful for the discovery of hierarchical clusters, and analysis of multiple factors providing insight in various real datasets including telecommunications churn.

Most windowing systems follow the independent overlapping windows approach, which emerged as an answer to the needs of early computer users. Due to advances in computers, display technology, and increased information needs, modern users demand more functionality from window management systems. We propose Elastic Windows with improved spatial layout and rapid multi-window operations as an alternative to current window management strategies. In this approach, multi-window operations are achieved by issuing operations on window groups hierarchically organized in a space-filling tiled layout similar to TreeMaps.1 Sophisticated multi-window operations have been developed to handle fast task-switching and to structure the work environment of users to their rapidly changing needs. We claim that these multi-window operations and the tiled spatial layout dynamics decrease the cognitive load on users by decreasing the number of window operations. This paper describes the Elastic Windows interface in detail and then presents a user study conducted to compare the performance of 12 users with Elastic Windows and traditional Independent Overlapping Windows. User performance was measured in terms of task environment setup, switching, and task execution for 2, 6, and 12 window situations. Elastic Windows users had statistically significantly faster performance for all tasks in 6 and 12 window situations. These results suggest promising possibilities for multiple window operations and hierarchical nesting, which can be applied to the next generation of tiled as well as overlapped window managers. © 1998 John Wiley & Sons, Ltd.

Most windowing systems follow the independent overlapping windows approach, which emerged as an answer to the needs of the 1980s' technology.Due to advances in computers and display technology, and increased information needs, modem users demand more functionality from window management systems.We proposed Elastic Windows with improved spatial layout and rapid multi-window operations as an alternative to current window management strategies for efficient personal role management [12].In this approach, multi-window operations are achieved by issurng operations on window groups hierarchically organized in a space-filling tiled layout.'l%is paper describes the Elastic Windows interface briefly and then presents a study comparing user performance with Elastic Windows and traditional window management techniques for 2, 6, and 12 window situations.Elastic Whdows users had statistically significantly faster performance for all 6 and 12 window situations, for task environment setup, task environment switching, and task execution.For some tasks there was a ten-fold speed-up in performanceThese results suggest promising possibilities for multiple window operations and hierarchical nesting, which can be applied to the next generation of tiled as well as overlapped window managers.

Article Elastic Windows: a hierarchical multi-window World-Wide Web browser Share on Authors: Eser Kandogan Department of Computer Science, Human-Computer Interaction Laboratory Department of Computer Science, Human-Computer Interaction LaboratoryView Profile , Ben Shneiderman Institute for Advanced Computer Studies, University of Maryland, College Park, MD Institute for Advanced Computer Studies, University of Maryland, College Park, MDView Profile Authors Info & Claims UIST '97: Proceedings of the 10th annual ACM symposium on User interface software and technologyOctober 1997 Pages 169–177https://doi.org/10.1145/263407.263541Published:01 October 1997 27citation762DownloadsMetricsTotal Citations27Total Downloads762Last 12 Months12Last 6 weeks4 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