Rosina O Weber

Rosina O Weber
Drexel University | DU

PhD

About

125
Publications
32,275
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,507
Citations

Publications

Publications (125)
Article
Full-text available
We present a scoping review of user studies in explainable artificial intelligence (XAI) entailing qualitative investigation. We draw on social science corpora to suggest ways for improving the rigor of studies where XAI researchers use observations, interviews, focus groups, and/or questionnaire tasks to collect qualitative data. We contextualize...
Presentation
Full-text available
Explainable AI tutorial discusses: 1) The third wave of AI: Explainable AI (XAI), Interpretable Machine Learning (IML) 2) Categories of XAI 3) Popular XAI methods 4) Categories of explanations 5) Evaluation of XAI
Preprint
Full-text available
Previous research in interpretable machine learning (IML) and explainable artificial intelligence (XAI) can be broadly categorized as either focusing on seeking interpretability in the agent's model (i.e., IML) or focusing on the context of the user in addition to the model (i.e., XAI). The former can be categorized as feature or instance attributi...
Preprint
Full-text available
The underlying hypothesis of knowledge-based explainable artificial intelligence is the data required for data-centric artificial intelligence agents (e.g., neural networks) are less diverse in contents than the data required to explain the decisions of such agents to humans. The idea is that a classifier can attain high accuracy using data that ex...
Presentation
Full-text available
Underspecification in machine learning has been blamed for lack of accuracy. Consider that there may be important aspects that connect features and instances that are not explicitly represented in data. Such limitation is obviously transferred to interpretable machine learning (IML) and explainable artificial intelligence (XAI) methods given they a...
Conference Paper
Full-text available
Various literature surveys state and confirm a rapid increase in research on explainable artificial intelligence (XAI) in recent years. One possible motivation for this change are legal regulations, including the general data protection regulation (GDPR) but also similar regulations outside of Europe. Another possible reason is the decreasing trust...
Article
Full-text available
This article describes how to utilize data-driven artificial intelligence (AI) to automate researcher assessment using data from profiling systems. We consider that a researcher assessment is done for a purpose and not divorced from a specific target placement. We formulate researcher assessment as a binary classification task, that is, a candidate...
Preprint
Full-text available
We present a focused analysis of user studies in explainable artificial intelligence (XAI) entailing qualitative investigation. We draw on social science corpora to suggest ways for improving the rigor of studies where XAI researchers use observations, interviews, focus groups, and/or questionnaires to capture qualitative data. We contextualize the...
Preprint
Explainable artificial intelligence (XAI) methods are currently evaluated with approaches mostly originated in interpretable machine learning (IML) research that focus on understanding models such as comparison against existing attribution approaches, sensitivity analyses, gold set of features, axioms, or through demonstration of images. There are...
Conference Paper
Full-text available
Neural network-based citation recommenders propose citations for a query article based on its latent relationships with corresponding recommended articles. The explanations for such citation recom-mendations can be accomplished in two ways: 1) Explain how the model works (i.e., internal), or 2) Explain the recommendation in the context of the user...
Article
Full-text available
In research funding, the challenge is to determine which proposers are more likely to succeed. Assessments based on the research proposal merit, or bibliometrics, have been adopted by managers of science and technology. Criticisms to such methods question that the different purposes of assessments are not considered, for instance, the purposes of f...
Presentation
Full-text available
Esta apresentação foi realizada durante o I Seminário de Avaliação de Políticas de CT&I, promovido pelo Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) e o Centro de Gestão e Estudos Estratégicos (CGEE), no eixo, Metodologias de avaliação e mensuração de impactos de programas, políticas e ações de CT&I. This presentation is a...
Conference Paper
Full-text available
This paper demonstrates how case-based reasoning (CBR) can be used for an explainable artificial intelligence (XAI) approach to justify solutions produced by an opaque learning method (i.e., target method), particularly in the context of unstructured textual data. Our general hypothesis is twofold: 1) There exists patterns in the relationship betwe...
Conference Paper
Full-text available
As part of our motivation to advance societal acceptance of and trust in explainable artificial intelligence (XAI)—namely, explainable case-based reasoning (XCBR) systems—we recognize the criticality of ascertaining how to properly approach explanation quality measurement. In this paper, we search the literature to facilitate decomposition of what...
Chapter
Full-text available
This paper demonstrates how case-based reasoning (CBR) can be used for an explainable artificial intelligence (XAI) approach to justify solutions produced by an opaque learning method (i.e., target method), particularly in the context of unstructured textual data. Our general hypothesis is twofold: (1) There exists patterns in the relationship betw...
Chapter
Full-text available
Digital change and scientific development have mutual implications. On one hand, science and technology development has been a major factor to digital change. On the other hand, the digital era has brought major changes to scientific knowledge production. First, there is a cyberinfrastructure—not only infrastructure for computing, but a major virtu...
Article
Purpose By establishing a conceptual path through the field of artificial intelligence for objectivistic knowledge artifacts (KAs), the purpose of this paper is to propose an extension to their design principles. The author uses these principles to deploy KAs for knowledge acquired in scientific processes, to determine whether these principles ste...
Conference Paper
Full-text available
Cyberinfrastructure refers to the computational infrastructure that supports the productivity and impact of scientific fields. Many scientific fields of study rely on science gateways such as HUBzero to build cyberinfrastructure portals that offer standard features such as databases, simulation, visualization, and grid tools. This article aims to a...
Conference Paper
Full-text available
Microtext is sparse and informal content typical in social media that is being widely used to study various facets of today’s society. This paper proposes the use of supplemental context to counteract the limitations imposed by the sparsity and the informality of microtext on the performance of word sense disambiguation (WSD). WSD relies on the sen...
Article
Full-text available
This work intends to combine domain ontology with natural language processing techniques to identify the sentiment behind judgments aiming to provide an explanation for such polarization. Also, it intends to use the Case-Based Reasoning strategy in order to learn from past reasonings (polarizations) so they can be used in new polarizations. Some st...
Conference Paper
Full-text available
We review the literature in search for attributes to characterize scholarly researchers, with a particular focus on collaborative work given its vast relevance and ubiquity. Our ultimate goal is to conduct studies to inform research related decisions, primarily focusing on researcher quality assessment. Recent efforts to design and maintain high-qu...
Conference Paper
Full-text available
Data generated across time may not be easily comparable in its original form thus potentially leading to results that may be perceived as unfair to some. We investigate quality assessment of scholarly researchers from their curricula vitae (CVs) for processes such as hiring, promotion, and grant funding. In previous work, we demonstrated that case-...
Conference Paper
Full-text available
This paper contributes, illustrates, and evaluates a purpose-oriented method to assess quality of researchers. It is in selection processes such as recruitment that assessing the quality of researchers becomes necessary. Because quality is fitness for use, we contend that assessing the quality of researchers depends on references of quality imposed...
Conference Paper
Full-text available
Users of sentiment analysis applications are interested in opinions of individual aspects (e.g., excellent mileage) of a target entity rather than in their polarity (e.g., 56 % are positive). This analysis is known as aspect-level sentiment analysis. In this paper, we use document-level polarization to learn patterns for contextual polarity, which...
Article
Full-text available
Guest editor’s introduction: special issue on case-based reasoning David B. Leake, Barry Smyth and Rosina Weber This Special Issue contains three articles providing samples of current case-based reasoning (CBR) research. Case-based reasoning systems perform problem-solving and interpretation based on a library of prior cases, called the case base....
Book
Full-text available
While it is relatively easy to record billions of experiences in a database, the wisdom of a system is not measured by the number of its experiences but rather by its ability to make use of them. Case-based rea­soning (CBR) can be viewed as experience mining, with analogical reasoning applied to problem–solution pairs. As cases are typically not id...
Conference Paper
Full-text available
The availability of new data sources presents both opportunities and challenges for the use of Case-based Reasoning to solve novel problems. In this paper, we describe the research challenges we faced when trying to reuse experiences of successful academic collaborations available online in descriptions of funded grant proposals. The goal is to rec...
Chapter
Full-text available
This chapter starts by providing a general view of advanced concepts in preparation for a deeper discussion of the elements explained in Part II. It starts by discussing the advanced aspect of the relationships between containers. The new concepts presented here are however of heterogeneous character in order to cover some aspects not yet presented...
Chapter
The chapter presents the basic notions used in CBR and necessary for understanding the remainder of this book. Case-based reasoning is a reasoning methodology for problem solving. It mainly relies on experiences in which problems were solved in the past. CBR reuses previous experiences to solve current, new problems. Problem solving experiences inc...
Chapter
Knowledge management (KM) concerns the proper allocation, coordination and planning of an organisation’s intellectual assets. Despite being an organisational problem, its solutions span multiple disciplines. The perspective adopted here is that of information technology, using the CBR methodology to perform KM tasks. The information technology pers...
Chapter
The chapter introduces case representations methods. The kind of case representation is decisive in all steps of the CBR methodology as all the process steps operate on case representations, hence they will also be mentioned in the subsequent chapters in Part II. It also introduces more complex representation methods that are later explained in Par...
Chapter
Chapter 9 is concerned with the reuse of a retrieved experience. Because the previous situation may not be exactly the same as the new problem, the retrieved solution may need to undergo some change. This is called adaptation. It allows us to reach not only the solutions that are directly stored in the case base but also those that are available by...
Chapter
This advanced chapter is intended for readers who have to deal with knowledge sources in textual format. There are two main strategies for addressing textual CBR. One is to convert any text into a machine readable form and manipulate it using methods discussed in the previous chapters. The other is to keep the original textual representation and ad...
Chapter
Chapter 11 examines life cycle development steps and how these steps, in addition to other methods, can help maintain the quality of a CBR system. The view of the performance of CBR systems is split into the initial situation, where the systems are built the first time and, the situations where systems have to be maintained. The chapter is addresse...
Chapter
The chapter extends the view of CBR as presented in Chap. 2, Basic CBR Elements, which adopts experiences of the type problem-solution. That standard view is now extended to experiences that can be of different natures. There is still a query problem and a solution, and a similarity measure. The similarity attempts to identify the relation of usefu...
Chapter
Chapter 14 has two parts. The first part of this chapter is addressed to readers who deal with complex objects and need advanced retrieval algorithms. First, we consider two advanced retrieval methods: Case Retrieval Nets and Fish and Shrink. Both require a special case representation form. In the case of retrieval nets, the queries and the cases a...
Chapter
Chapter 22, Basic Formal Definitions and Methods, deals with formal aspects mentioned earlier in the book. They are logic and truth, information content, and utility. The purpose of these topics is as follows: (a) Is the provided information correct? (b) How much information is provided? (c) Is the information useful? In this chapter we provide the...
Chapter
Chapter 23 is concerned with areas that have a close relation to CBR. CBR systems have more or less close connections to other methods. It depends on the problem and the interests of the reader on how to make a choice of the methods. The intention of this chapter is twofold: First, finding out under which circumstances one should choose which metho...
Chapter
Chapter 6 is intended to give the reader a first advice on which similarity measure is suitable for which task, when designing a CBR system. For this purpose, it contains the basic material necessary for understanding and building simple similarity measures. Similarity assessment is a substep of the retrieve process step and is responsible for iden...
Chapter
Chapter 7 complements Chap. 6 in presenting similarity topics, extending similarity measures to more complex case representations. We focus mainly on structural similarities tailored to complex object representations. We follow case representations given in Chap. 5, Case Representations, and include additional ones. This is intended as a help in ch...
Chapter
This chapter is addressed to readers who work with and are interested in time series of sensor data and in the recognition of spoken language. We consider representations of cases and queries in such formats. We make use of signal processing but do not discuss detailed signal questions in speech. Rather, we are interested in problems concerning use...
Chapter
This advanced chapter is addressed to readers who are interested in foundational aspects, formal aspects, subjectivity, functional dependency, and other special additional problems. The general aspects of similarities are extended and deepened. We introduce a semantics for similarity measures allowing us to define correctness for similarity-based c...
Chapter
This advanced chapter addresses topics that are of particular concern to readers dealing with images occurring as queries or in the representation of cases. In domains like medicine or geography, images render large amounts of important data that are easily accessible. First, the representation of the objects is introduced. Following the local-glob...
Chapter
Chapter 8 presents basic retrieval algorithms. The contents in this chapter will help a designer determine how to efficiently retrieve useful cases, what depends on the complexity of case representation and similarity assessment. It includes methods such as sequential retrieval, two-level retrieval, and retrieval methods with more complex indexing,...
Chapter
Chapter 15 discusses relations between CBR and some uncertainty concepts. The main topics are rough sets and fuzzy sets. These concepts occur frequently in applications. Of special interest is the relation to similarity. The rough set concept provides a method of describing uncertain results in a user-controlled way. Fuzzy sets and their logical de...
Chapter
The first part of this chapter is basic and is of interest to readers who want to build or maintain a CBR system. The second part provides background knowledge. It deals with revising the methods if something is definitely wrong and with improving CBR systems by machine learning if the results are weak. In this chapter methods for evaluating, revis...
Chapter
This advanced chapter discusses topics in probabilities of interest to CBR. It is devoted to readers who deal in their applications with stochastic phenomena. Some basic knowledge about probabilities is required. We discuss that the connections between similarities and probabilities are manifold. There are two directions: Probabilities give rise to...
Chapter
This chapter deals with situations where the user does not present a complete problem in the first interaction with a CBR system. In conversational CBR systems knowledge is manipulated via the management of conversations. Conversations are related to all knowledge containers, but the most important one is the vocabulary container. This is because t...
Chapter
In this chapter examples support and complement the understanding of the previous chapters. In this chapter examples of major problem areas as classification, diagnosis, prediction, call centres, etc. are briefly introduced. We mainly discuss how example applications can be formulated in a form suitable for CBR. The examples show the variety of app...
Article
Full-text available
Finding collaborators to engage in academic research is a challenging task, especially when the collaboration is multidisciplinary in nature and collaborators are needed from different disciplines. This paper uses evidence of successful multidisciplinary collaborations, funded proposals, in a novel way: as an input for a method of recommendation of...
Article
Full-text available
A representation for scientific knowledge can enable the computational manipulation of scientific contents. The goal is to facilitate the construction of deliverables in support of education. These learning products can range in a variety of dimensions: in format from textual to computational; in educational content from scientific articles to text...
Article
Full-text available
As research collaboration in academia has increased over the past century, we now see a similar increase in collaboration across disciplinary boundaries, particularly in the field of library and information science, a field that has long been viewed as highly multidisciplinary. This increase is driven by the pressures on faculty seeking career adva...
Article
Full-text available
Working Capital Management is concerned with the short term decisions regarding current assets and current liabilities. Although being related with the short term, the decisions must take into account the overall strategy of the firm and how effectively they are implemented at the operational level. The satisfaction of the goals and the measurement...
Conference Paper
Full-text available
Expert locators and repository-based knowledge management systems (KMS) are different architectures proposed to perform different kinds of knowledge management functions. While expert locators can recommend an expert to perform a task, repository-based KMS can share a learned strategy to solve a given problem In this paper, we describe a framework...
Article
New scientific fields often grow from the intersections and boundaries between existing disciplines. Before these new domains are well established, with important authors and key journals, publications are usually mixed into the journals and conferences of the parent domains, making it difficult to identify core works. This paper presents a validat...
Article
Full-text available
This article describes a knowledge management (KM) approach conceived from countermeasures targeted at addressing failure factors suggested in the literature. In order to counteract failure factors, the approach combines the technology of knowledge-based KM systems, with the flexibility and understanding of knowledge facilitators, and the processes...
Article
Increasingly, knowledge is being created and applied on the move by knowledge workers. In this paper, we discuss the concept of nomadic context-aware Knowledge Management Systems (KMS). In nomadic context-aware KMS, computing becomes inseparable from the environment and work of the knowledge worker, and they serve to support their knowledge intensi...
Article
Full-text available
Knowledge artifacts are units retained in repository-based knowledge management systems (KMS). When adopting cases as the representation for knowledge artifacts in KMS, one of the benefits is the automated construction of reports.
Conference Paper
Full-text available
CAMRA (Center for Advancing Microbial Risk Assessment) gathers a community of scientists that investigate several stages in the life cycle of biological agents of concern. This paper describes the knowledge management (KM) approach adopted for CAMRA’s community of scientists. The approach includes knowledge facilitators, a web- and repository-based...
Article
Full-text available
Case-based reasoning (CBR) is a reasoning methodology that relies on previous experiences, making it well suited to various real world application domains. When we use CBR to solve real world problems, its inherent uncertainty tends to propagate and may be detrimental to the system's quality. Consequently, the quality of a CBR system's outcome tend...
Article
Full-text available
The search for an online product that matches e-shoppers’ needs and preferences can be frustrating and time-consuming. Browsing large lists arranged in tree-like structures demands focused attention from eshoppers. Keyword search often results in either too many useless items (low precision) or few or none useful ones (low recall). This can cause p...
Conference Paper
Full-text available
The configuration of a computational intelligence (CI) method is responsible for its intelligence (e.g. tolerance, flexibility) as well as its accuracy. In this paper, we investigate how to automatically improve the performance of a CI method by finding alternate configuration parameter values that produce more accurate results. We explore this by...
Article
Full-text available
This commentary provides a definition of textual case-based reasoning (TCBR) and surveys research contributions according to four research questions. We also describe how TCBR can be distinguished from text mining and information retrieval. We conclude with potential directions for TCBR research.
Article
Full-text available
This commentary describes two core knowledge management approaches that applied case-based reasoning as a methodological foundation for organizational systems managing experience. These research projects illustrate the presence of knowledge management in case-based reasoning by focusing on the dualism between case-based reasoning and organizational...
Article
Full-text available
The aim of this commentary is to discuss the contribution of soft computing—a consortium of fuzzy logic, neural network theory, evolutionary computing, and probabilistic reasoning—to the development of case-based reasoning (CBR) systems. We will describe how soft computing has been used in case representation, retrieval, adaptation, reuse, and case...
Conference Paper
Full-text available
This paper describes how CBR can be used to compare, reuse, and adapt inductive models that represent complex systems. Complex systems are not well understood and therefore require models for their manipulation and understanding. We propose an approach to address the challenges for using CBR in this context, which relate to finding similar inductiv...
Conference Paper
Full-text available
This paper describes an integrated approach for detecting inconspicuous contents in text. Inconspicuous contents can be an opinion or goal that may be disguised in some way to mislead automated methods but keeps a clear message for humans (e.g., terrorist sites). Our methodology hypothesizes that patterns that convey inconspicuous contents can be e...
Article
Full-text available
This paper explores a method to algorithmically distinguish case-specific facts from potentially reusable or adaptable elements of cases in a textual case-based reasoning (TCBR) system. In the legal domain, documents often contain casespecific facts mixed with case-neutral details of law, precedent, conclusions the attorneys reach by applying their...
Conference Paper
Full-text available
Textual case-based reasoning (TCBR) provides the ability to reason with domain-specific knowledge when experiences exist in text. Ideally, we would like to find an inexpensive way to automatically, efficiently, and accurately represent textual documents as cases. One of the challenges, however, is that current automated methods that manipulate text...
Conference Paper
Full-text available
When events such as severe weather or congestion interfere with the normal flow of air traffic, air traffic controllers may implement "plays" that reroute one or more traffic flows. Currently, plays are assessed and selected based on controllers’ experience using the National Playbook, a collection of plays that have worked in the past. This paper...
Conference Paper
Full-text available
Computer systems do not learn from previous experiences unless they are designed for this purpose. Computational intelligence systems (CIS) are inherently capable of dealing with imprecise contexts, creating a new solution in each new execution. Therefore, every execution of a CIS is valuable to be learned. We describe an architecture for designing...
Article
Full-text available
A learned lesson, in the context of a pre-defined organizational process, summarizes an experience that should be used to modify that process, under the conditions for which that lesson applies. To promote lesson reuse, many organizations employ lessons learned processes, which define how to collect, validate, store, and disseminate lessons among t...
Article
Full-text available
A reversible outcome is one that can be changed. For example, the failure of an ongoing project may be avoided if certain actions are taken, while an outcome such as the path of a hurricane cannot be changed under current knowledge. The major benefit of predicting reversible outcomes resides in the possibility to avoid unwanted results. For this pu...
Article
Full-text available
NutriGenomics is the bioscience that links the way nutrients and other dietary components shape genetic activity. It builds on the success of Human Genome Project by applying systems biology methods to explain how the molecular components of food, supplements and pharmaceuticals dynamically influence and shape the activity of genomic subsystems, wh...
Conference Paper
Full-text available
Monitored distribution (MD) is a case-based approach for proactive knowledge distribution. MD allows the dissemination of knowledge artifacts in a just-in-time fashion in the context of its applicable targeted processes. In MD, knowledge artifacts are retrieved when they are applicable to the task in which a user is currently engaged. We define MD'...
Conference Paper
Full-text available
Case-based reasoning is a flexible methodology to manage software development related tasks. However, when the reasoner’s task is prediction, there are a number of different CBR techniques that could be chosen to address the characteristics of a dataset. We examine several of these techniques to assess their accuracy in predicting software developm...
Article
Full-text available
Lessons learned systems are knowledge management solutions that serve the purpose of capturing, storing, disseminating and sharing an organization's verified lessons. In this paper we propose a two-step categorization method to support the design of intelligent lessons learned systems. The first step refers to the categories of the lessons learned...
Article
Full-text available
Intelligent Jurisprudence Research (IJR) is a concept that consists in performing jurisprudence research with a computational tool that employs Artificial Intelligence (AI) techniques. Jurisprudence research is the search employed by judicial professionals when seeking for past legal situations that may be useful to a legal activity.
Article
Full-text available
We introduce LET (Lesson Elicitation Tool), which uses domain and linguistic knowledge to guide users during their submission of lessons learned. LET can detect a user's need for instructions and disambiguates expressions while collecting taxonomic domain knowledge.
Article
Full-text available
Lessons learned systems (LLS) are a common knowledge management (KM) initiative among the American government agencies (e.g., Department of Defense (DOD), Department of Energy (DOE), NASA). An effective lessons learned (LL) process can substantially improve decision processes, thus representing an essential chapter in a knowledge-sharing digital go...