Yolanda Gil

AAAI Conference 2017 Conference Paper

Towards Continuous Scientific Data Analysis and Hypothesis Evolution

• Yolanda Gil
• Daniel Garijo
• Varun Ratnakar
• Rajiv Mayani
• Ravali Adusumilli
• Hunter Boyce
• Arunima Srivastava
• Parag Mallick

Scientific data is continuously generated throughout the world. However, analyses of these data are typically performed exactly once and on a small fragment of recently generated data. Ideally, data analysis would be a continuous process that uses all the data available at the time, and would be automatically re-run and updated when new data appears. We present a framework for automated discovery from data repositories that tests user-provided hypotheses using expert-grade data analysis strategies, and reassesses hypotheses when more data becomes available. Novel contributions of this approach include a framework to trigger new analyses appropriate for the available data through lines of inquiry that support progressive hypothesis evolution, and a representation of hypothesis revisions with provenance records that can be used to inspect the results. We implemented our approach in the DISK framework, and evaluated it using two scenarios from cancer multi-omics: 1) data for new patients becomes available over time, 2) new types of data for the same patients are released. We show that in all scenarios DISK updates the confidence on the original hypotheses as it automatically analyzes new data.

AAAI Conference 2016 Conference Paper

Teaching Big Data Analytics Skills with Intelligent Workflow Systems

• Yolanda Gil

We have designed an open and modular course for data science and big data analytics using a workflow paradigm that allows students to easily experience big data through a sophisticated yet easy to use instrument that is an intelligent workflow system. A key aspect of this work is the use of semantic workflows to capture and reuse end-to-end analytic methods that experts would use to analyze big data, and the use of an intelligent workflow system to elaborate the workflow and manage its execution and resulting datasets. Through the exposure of big data analytics in a workflow framework, students will be able to get first-hand experiences with a breadth of big data topics, including multi-step data analytic and statistical methods, software reuse and composition, parallel distributed programming, high-end computing. In addition, students learn about a range of topics in AI, including semantic representations and ontologies, machine learning, natural language processing, and image analysis.

KER Journal 2011 Journal Article

Interactive knowledge capture in the new millennium: how the Semantic Web changed everything

• Yolanda Gil

Abstract The Semantic Web has radically changed the landscape of knowledge acquisition research. It used to be the case that a single user would edit a local knowledge base, that the user would have domain expertise to add to the system, and that the system would have a centralized knowledge base and reasoner. The world surrounding knowledge-rich systems changed drastically with the advent of the Web, and many of the original assumptions were no longer a given. Those assumptions had to be revisited and addressed in combination with new challenges that were put forward. Knowledge-rich systems today are distributed, have many users with different degrees of expertise, and integrate many shared knowledge sources of varying quality. Recent work in interactive knowledge capture includes new and exciting research on collaborative knowledge sharing, collecting knowledge from Web volunteers, and capturing knowledge provenance.

IS Journal 2011 Journal Article

Wings: Intelligent Workflow-Based Design of Computational Experiments

• Yolanda Gil
• Varun Ratnakar
• Jihie Kim
• Pedro Gonzalez-Calero
• Paul Groth
• Joshua Moody
• Ewa Deelman

Describes the Wings intelligent workflow system that assists scientists with designing computational experiments by automatically tracking constraints and ruling out invalid designs, letting scientists focus on their experiments and goals.

AAAI Conference 2008 Conference Paper

Automating To-Do Lists for Users: Interpretation of To-Dos for Selecting and Tasking Agents

• Yolanda Gil

To-do lists have been found to be the most popular personal information management tools, yet there is no automated system to interpret and act upon them when appropriate on behalf of the user. Automating to-do lists is challenging, not only because they are specified as free text but also because most items contain abbreviated tasks, many do not specify an action to be performed, and often refer to unrelated (personal) items. This paper presents our approach and an implemented system to process to-do list entries and map them to tasks that can be automated for the user by a set of agents. Since the format of to-do entries is not very amenable to natural language processing tools that can parse and create a structured interpretation, our approach is to exploit paraphrases of the target tasks that the agents can perform and that specify how the free-text maps to the task arguments. As users manually assign to-do to agents for automation, our system improves its performance by learning new paraphrases. We show an evaluation of our approach in a corpus of 2100 to-do entries collected from users of an office assistant multi-agent system.

IJCAI Conference 2003 Conference Paper

Proactive Dialogue for Interactive Knowledge Capture

• Jihie Kim
• Yolanda Gil

Current tools for interactive knowledge capture have little or no learning aptitude. They are mostly oblivious to the process or strategy that the user may be following in entering new knowledge, unaware of their progress during a session, and ignorant of typical skills expected from a good student. We present an approach to make acquisition interfaces more proactive by extending them with: 1) goals that represent what remains to be learned, 2) strategies to achieve these goals and acquire further knowledge, and 3) awareness of the current status of the body of knowledge learned. The resulting interaction shows that the system is aware of its progress towards acquiring the new knowledge, and moves forward by understanding what acquisition goals and strategies to pursue.

ICAPS Conference 2003 Conference Paper

The Role of Planning in Grid Computing

• Jim Blythe
• Ewa Deelman
• Yolanda Gil
• Carl Kesselman
• Amit Agarwal
• Gaurang Mehta
• Karan Vahi

Grid computing gives users access to widely distributed networks of computing resources to solve large-scale tasks such as scientific computation. These tasks are defined as standalone components that can be combined to process the data in various ways. We have implemented a planning system to generate task workflows for the Grid automatically, allowing the user to specify the desired data products in simple terms. The planner uses heuristic control rules and searches a number of alternative complete plans in order to find a high-quality solution. We describe an implemented test case in gravitational wave interferometry and show how the planner is integrated in the Grid environment. We discuss promising future directions of this work. We believe AI planning will play a crucial role in developing complex application workflows for the Grid.

AAAI Conference 1999 Conference Paper

Deriving Expectations to Guide Knowledge Base Creation

• Jihie Kim
• Yolanda Gil
• University of Southern California

Successful approaches to developing knowledge acquisition tools useexpectationsof whatthe userhasto add ormay want to add, basedon how new knowledgefits within a knowledge base that already exists. When a knowledge base is first created or undergoes significant extensions and changes, these tools cannot provide much support. This paper presents an approach to creating expectations when a new knowledge base is built, and describes a knowledge acquisition tool that we implemented using this approach that supports users in creating problem-solving knowledge. As the knowledge base grows, the knowledge acquisition tool derives more frequent and more reliable expectations that result from enforcing constraints in the knowledge representation system, looking for missing pieces of knowledge in the knowledge base, and working out incrementally the inter-dependencies among the different components of the knowledgebase. Our preliminary evaluations show a thirty percent time savings during knowledge acquisition. Moreover, by providing tools to support the initial phases of knowledge base development, many mistakes are detected early on and even avoided altogether. We believe that our approach contributes to improving the quality of the knowledgeacquisition processand of the resulting knowledge-based systems as well.

AAAI Conference 1999 Conference Paper

Designing Scripts to Guide Users in Modifying Knowledge-Based Systems

• Marcelo Tallis
• Yolanda Gil
• University of Southern California

Knowledge Acquisition (KA)Scripts capture typical modificationsequencesthat users follow when they modify knowledgebases. KAtools can use these Scripts to guide users in makingthese modifications, ensuringthat they follow all the ramifications of the changeuntil it is completed. This paper describes our approachto design, develop, and organize a library of KAScripts. Wereport the results of three different analysis to develop this library, including a detailed study of actual modification scenarios in two knowledgebases. In addition to identifying a good numberof KA Scripts, wefounda set of useful attributes to describe and organize the KAScripts. These attributes allowus to analyzethe size of the library and generate newKAScripts in a systematic way. Wehaveimplemented a portion of this library and conducted twodifferent studies to evaluate it. The result of this evaluation showeda 15 to 52 percent time savings in modifying knowledgebases andthat the library included relevant and useful KA Scripts to assist users in realistic settings.

AAAI Conference 1997 Conference Paper

A Script-Based Approach to Modifying Knowledge Bases

• Yolanda Gil

Our goal is to build knowledge acquisition tools that support users in modifying knowledge-based systems. These modifications may require several individual changes to various components of the knowledge base, which need to be carefully coordinated to prevent users from leaving the knowledge-based system in an unusable state. This paper describes an approach to building knowledge acquisition tools which capture knowledge about commonly occurring modification sequences and support users in completing the modifications they start. These sequences, which we call ICA Scripts, relate individual changes and the effects that they have on the knowledge base. We discuss our experience in designing and compiling a library of KA Scripts. We also describe the implementation of a tool that uses them and our preliminary evaluations that demonstrate their usability.

AAAI Conference 1996 Conference Paper

Explicit Representations of Problem-Solving Strategies to Support Knowledge Acquisition

• Yolanda Gil

Role-limiting approaches support knowledge acquisition (KA) by centering knowledge base construction on common types of tasks or domainindependent problem-solving strategies. Within a particular problem-solving strategy, domaindependent knowledge plays specific roles. A KA tool then helps a user to fill these roles. Although role-limiting approaches are useful for guiding KA, they are limited because they only support users in filling knowledge roles that have been built in by the designers of the KA system. EX- PECT takes a different approach to KA by representing problem-solving knowledge explicitly, and deriving from the current knowledge base the knowledge gaps that must be resolved by the user during KA. This paper contrasts role-limiting approaches and EXPECT’ s approach, using the propose-and-revise strategy as an example. EX- PECT not only supports users in filling knowledge roles, but also provides support in making other modifications to the knowledge base, including adapting the problem-solving strategy. EXPECT’ s guidance changes as the knowledge base changes, providing a more flexible approach to knowledge acquisition. This work provides evidence supporting the need for explicit representations in building knowledge-based systems.

ICAPS Conference 1996 Conference Paper

Planning Experiments: Resolving Interactions between Two Planning Spaces

• Yolanda Gil

Learning from experimentation allows a system to acquire planning domain knowledge by correcting its knowledge when an action execution fails. Experiments are designed and planned to bring the world to a state where a hypothesis (e. g., that an operator is missing a precondition) can be tested. When planning an experiment, the planner must take into account the interactions between the execution of the main plan and the execution of the experiment plans, since after the experiment itmust continue to carry on its main task. In order for planners to work in such environments where they can be given several tasks, they must take into account the interactions between them. A usual assumption in current planning systems is that they are given a single task (or set of goals to achieve). However, a plan that may seem adequate for a task in isolation may make other tasks harder (or even impossible) to achieve. Different tasks may compete for resources, execute irreversible actions that make other tasks unachievable, or set the world in undesirable states. This paper discusses what these interactions are and presents how the problem was adressed in EXPO, an implemented system that acquires domain knowledge for planning through experimentation.

AAAI Conference 1994 Conference Paper

Knowledge Refinement in a Reflective Architecture

• Yolanda Gil

A knowledge acquisition tool should provide a user with maximum guidance in extending and debugging a knowledge base, by preventing inconsistencies and knowledge gaps that may arise inadvertently. Most current acquisition tools are not very flexible in that they are built for a predetermined inference structure or problem-solving mechanism, and the guidance they provide is specific to that inference structure and hardcoded by their designer. This paper focuses on EXPECT, a reflective architecture that supports knowledge acquisition based on an explicit analysis of the structure of a knowledge-based system, rather than on a fixed set of acquisition guidelines. EXPECT'S problem solver is tightly integrated with LOOM, a state-of-the-art knowledge representation system. Domain facts and goals are represented declaratively, and the problem solver keeps records of their functionality within the task domain. When the user corrects the system’ s knowledge, EXPECT tracks any possible implications of this change in the overall system and cooperates with the user to correct any potential problems that may arise. The key to the flexibility of this knowledge acquisition tool is that it adapts its guidance as the knowledge bases evolve in response to changes introduced by the user.

ICML Conference 1994 Conference Paper

Learning by Experimentation: Incremental Refinement of Incomplete Planning Domains

• Yolanda Gil

ICML Conference 1993 Conference Paper

Efficient Domain-Independent Experimentation

• Yolanda Gil

AIJ Journal 1989 Journal Article

Explanation-based learning:A problem solving perspective

• Steven Minton
• Jaime G. Carbonell
• Craig A. Knoblock
• Daniel R. Kuokka
• Oren Etzioni
• Yolanda Gil

This article outlines explanation-based learning (EBL) and its role in improving problem solving performance through experience. Unlike inductive systems, which learn by abstracting common properties from multiple examples, EBL systems explain why a particular example is an instance of a concept. The explanations are then converted into operational recognition rules. In essence, the EBL approach is analytical and knowledge-intensive, whereas inductive methods are empirical and knowledge-poor. This article focuses on extensions of the basic EBL method and their integration with the prodigy problem solving system. prodigy's EBL method is specifically designed to acquire search control rules that are effective in reducing total search time for complex task domains. Domain-specific search control rules are learned from successful problem solving decisions, costly failures, and unforeseen goal interactions. The ability to specify multiple learning strategies in a declarative manner enables EBL to serve as a general technique for performance improvement. prodigy's EBL method is analyzed, illustrated with several examples and performance results, and compared with other methods for integrating EBL and problem solving.