Ming Cheng, Jiaying Gong, Chenhan Yuan, William A Ingram, Edward Fox, Hoda Eldardiry. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Electronic theses and dissertations (ETDs) have been proposed, advocated, and generated for more than 25 years. Although ETDs are hosted by commercial or institutional digital library repositories, they are still an understudied type of scholarly big data, partially because they are usually longer than conference and journal papers. Segmenting ETDs will allow researchers to study sectional content. Readers can navigate to particular pages of interest, to discover and explore the content buried in these long documents. Most existing frameworks on document page classification are designed for classifying general documents, and perform poorly on ETDs. In this paper, we propose ETDPC. Its backbone is a two-stream multimodal model with a cross-attention network to classify ETD pages into 13 categories. To overcome the challenge of imbalanced labeled samples, we augmented data for minority categories and employed a hierarchical classifier. ETDPC outperforms the state-of-the-art models in all categories, achieving an F1 of 0.84 -- 0.96 for 9 out of 13 categories. We also demonstrated its data efficiency. The code and data can be found on GitHub (https://github.com/lamps-lab/ETDMiner/tree/master/etd_segmentation).

Traditional archival practices for describing electronic theses and dissertations (ETDs) rely on broad, high-level metadata schemes that fail to capture the depth, complexity, and interdisciplinary nature of these long scholarly works. The lack of detailed, chapter-level content descriptions impedes researchers’ ability to locate specific sections or themes, thereby reducing discoverability and overall accessibility. By providing chapter-level metadata information, we improve the effectiveness of ETDs as research resources. This makes it easier for scholars to navigate them efficiently and extract valuable insights. The absence of such metadata further obstructs interdisciplinary research by obscuring connections across fields, hindering new academic discoveries and collaboration. In this paper, we propose a machine learning and AI-driven solution to automatically categorize ETD chapters. This solution is intended to improve discoverability and promote understanding of chapters. Our approach enriches traditional archival practices by providing context-rich descriptions that facilitate targeted navigation and improved access. We aim to support interdisciplinary research and make ETDs more accessible. By providing chapter-level classification labels and using them to index in our developed prototype system, we make content in ETD chapters more discoverable and usable for a diverse range of scholarly needs. Implementing this AI-enhanced approach allows archives to serve researchers better, enabling efficient access to relevant information and supporting deeper engagement with ETDs. This will increase the impact of ETDs as research tools, foster interdisciplinary exploration, and reinforce the role of archives in scholarly communication within the data-intensive academic landscape.

We developed a deep learning approach to detect and segment nuclear pores in large high-resolution 3D FIB-SEM images. The approach extracts small blocks containing nuclear pore regions from different parts of a cell nucleus and applies data augmentation techniques combined with Random Block Sampling (RBS) to optimize training. By fine tuning the neural network and reducing the batch size to one, we enhanced the model’s ability to learn from individual image features, leading to more precise detection. Our Dynamic Cyclical Data Augmentation (DCDA) adapts in real-time with an automatic termination feature, reducing augmentation time by over 97.4% in our experiments, significantly improving efficiency without compromising model performance. Experimental results consistently demonstrate that our approach can quickly and accurately identify nuclear pores, achieving F1 scores from 0.7119 to 0.8555, with precision ranging from 0.7379 to 0.8516. Training time was also significantly reduced. These substantial improvements in both accuracy and efficiency underscore the method’s effectiveness for large-scale cellular analysis.

As research institutions increasingly commit to supporting the United Nations’ Sustainable Development Goals (SDGs), there is a pressing need to accurately assess their research output against these goals. Current approaches, primarily reliant on keyword-based Boolean search queries, conflate incidental keyword matches with genuine contributions, reducing retrieval precision and complicating benchmarking efforts. This study investigates the application of autoregressive Large Language Models (LLMs) as evaluation agents to identify relevant scholarly contributions to SDG targets in scholarly publications. Using a dataset of academic abstracts retrieved via SDG-specific keyword queries, we demonstrate that small, locally-hosted LLMs can differentiate semantically relevant contributions to SDG targets from documents retrieved due to incidental keyword matches, addressing the limitations of traditional methods. By leveraging the contextual understanding of LLMs, this approach provides a scalable framework for improving SDG-related research metrics and informing institutional reporting.

Metadata is crucial for the accessibility, interoperability, and long-term usability of digital objects such as Electronic Theses and Dissertations (ETDs). In large-scale academic repositories, poor metadata quality can significantly impede the discovery and use of resources. This study addresses persistent issues of incomplete and inconsistent ETD metadata collected from U.S. university libraries. However, directly applying machine learning-based error detection and correction models may introduce unwanted errors due to the imperfection of these models. We propose an ETD metadata improvement system (ETDMIS) that mitigates the problem by integrating metadata validation and a version control mechanism. Our system was applied to a dataset of 100,000 U.S. ETDs, resulting in substantial improvements in metadata quality. Scalability was demonstrated by processing the entire dataset efficiently. The original and the enhanced metadata for the 100,000 ETDs are publicly accessible at https://github.com/lamps-lab/ETDMiner/tree/master/Meta100K.

Theses and dissertations record the work of graduate students and are typically a requirement at the culmination of the graduate degree. Thus, they contain important information that reflects a graduate student’s exploration of their research topic. Although print submission was commonplace early on, most universities now require students to submit an electronic version. The electronic document referred to as an ETD henceforth has become the primary way of submitting, storing, and distributing graduate work. Millions of such documents have been created in the past two decades. They are maintained and stored by university libraries, digital repositories, and other academic publishing companies. These online repositories have increased access to such documents. Nonetheless, these documents fail to meet the needs of researchers, who find it challenging to find and access knowledge from such long documents. The worldwide ETD collection has increased in volume to become what is known as ‘scholarly big data’. Apart from the text body, these documents contain a myriad of other pieces of knowledge like tables, figures, definitions, literature reviews, and references. There is a growing demand amongst researchers across various domains to make this collection of scholarly documents more computationally driven. We use ideas from natural language processing, information retrieval, and machine learning to excavate knowledge from this rich information source. In this paper, we examine some of the challenges we face, identify some key areas of exploration, and discuss our methods to mitigate the challenges.

In this work, we report our progress on building a collection containing over 450k Electronic Theses and Dissertations (ETDs), including full-text and metadata. Our goal is to close the gap of accessibility between long text and short text documents, and to create a new research opportunity for the scholarly community. For that, we developed an ETD Ingestion Framework (EIF) that automatically harvests metadata and PDFs of ETDs from university libraries. We faced multiple challenges and learned many lessons during the process, that led to proposed solutions to overcome/mitigate the limitations of the current data. We also described the data that we have collected. We hope our methods will be useful for building similar collections from university libraries and that the data can be used for research and education.

Generative Adversarial Networks (GANs) can accurately model complex multi-dimensional data and generate realistic samples. However, due to their implicit estimation of data distributions, their evaluation is a challenging task. The majority of research efforts associated with tackling this issue were validated by qualitative visual evaluation. Such approaches do not generalize well beyond the image domain. Since many of those evaluation metrics are proposed and bound to the vision domain, they are difficult to apply to other domains. Quantitative measures are necessary to better guide the training and comparison of different GANs models. In this work, we leverage Siamese neural networks to propose a domain-agnostic evaluation metric: (1) with a qualitative evaluation that is consistent with human evaluation, (2) that is robust relative to common GAN issues such as mode dropping and invention, and (3) does not require any pretrained classifier. The empirical results in this paper demonstrate the superiority of this method compared to the popular Inception Score and are competitive with the FID score.

Multiple-display environments (MDEs) have promise in helping co-located sensemaking tasks by supporting searching, organizing, and discussion tasks. Co-located sensemaking occurs when two or more sensemakers forage for useful information within a dataset, creating and leveraging knowledge structures individually and together. Group territories in MDEs support communicating and assembling findings, but questions remain regarding how to best represent individual sensemaking efforts in the group territory to support the sensemaking collaboration. This paper empirically examines exploration of a large Twitter dataset using three group territory paradigms: parallel, connected, and merged. Results reveal that merging group work increases task complexity while separating individuals' sections in the group territory supports monitoring, and more interactions are performed when individual work is not connected in the group territory.

The recent increase in online user generated content (UGC) has led to the availability of a large number of posts about products and services. Often, these posts contain complaints that the consumers purchasing the products and services have. However, discovering and summarizing product defects and the related knowledge from large quantities of user posts is a difficult task. Traditional aspect opinion mining models, that aim to discover the product aspects and their corresponding opinions, are not sufficient to discover the product defect information from the user posts. In this paper, we propose the Product Defect Latent Dirichlet Allocation model (PDLDA), a probabilistic model that identifies domain-specific knowledge about product issues using interdependent three-dimensional topics: Component, Symptom, and Resolution. A Gibbs sampling based inference method for PDLDA is also introduced. To evaluate our model, we introduce three novel product review datasets. Both qualitative and quantitative evaluations show that the proposed model results in apparent improvement in the quality of discovered product defect information. Our model has the potential to benefit customers, manufacturers, and policy makers, by automatically discovering product defects from online data.

Arabic news articles in electronic collections are difficult to study. Browsing by category is rarely supported. Although helpful machine‐learning methods have been applied successfully to similar situations for English news articles, limited research has been completed to yield suitable solutions for Arabic news. In connection with a Qatar National Research Fund (QNRF)‐funded project to build digital library community and infrastructure in Qatar, we developed software for browsing a collection of about 237,000 Arabic news articles, which should be applicable to other Arabic news collections. We designed a simple taxonomy for Arabic news stories that is suitable for the needs of Qatar and other nations, is compatible with the subject codes of the International Press Telecommunications Council, and was enhanced with the aid of a librarian expert as well as five Arabic‐speaking volunteers. We developed tailored stemming (i.e., a new Arabic light stemmer called P‐Stemmer) and automatic classification methods (the best being binary Support Vector Machines classifiers) to work with the taxonomy. Using evaluation techniques commonly used in the information retrieval community, including 10‐fold cross‐validation and the Wilcoxon signed‐rank test, we showed that our approach to stemming and classification is superior to state‐of‐the‐art techniques.

Analysis of structural properties and dynamics of networks is currently a central topic in many disciplines including Social Sciences, Biology and Business. CINET, a cyber infrastructure for such studies, introduced the concept of supporting network analysis as a service. The basic idea is to allow experts in various disciplines to focus on obtaining domain-specific insights from the results of network analyses instead of worrying about programming details and allocation of computational resources needed to carry out the analyses. A basic version of CINET was released in May 2012. This paper discusses CINET 2.0, a significantly enhanced version that supports complex network analyses through a web portal. CINET 2.0 has already been used for teaching courses related to Network Science at several US universities. In this paper, we discuss how CINET 2.0 significantly extends CINET 1.0 through enhancements to some components and the addition of new components.

As computing moves into its sixth decade as a formal field of study, a great deal of attention is focused on the breadth of relationships between computing and other disciplines. Descriptors of this relationship include computing in context, computing with a purpose, computing + X, computational X, interdisciplinary computing, multi-disciplinary computing, and trans-disciplinary computing. Regardless of the descriptor, the goals for exploring the relationship remain the same: students need to understand how the power of computing informs and shapes ideas throughout the academy and society. They need to learn appropriate concepts in computing and to see these applied to other areas of study. This learning will be most effective when approached through active participation of students, as they tackle questions and problems coming from society's ever-growing and ever-changing reliance on computing devices and modes of interaction. Our NSF-funded project, titled Computing in Context, fosters interdisciplinary computing education by creating exemplars of curricular materials, testing these materials in undergraduate computing courses, and disseminating the materials through computingportal.org. We focus our effort on four areas of applied computing: computational linguistics, computing and music intelligence and security informatics, and web science. Our materials will be designed for inquiry-based learning, whether this occurs in lecture as a brief conversation between pairs of students or in lab as loosely formulated problem or scenario or as a multi-week, team-based project.

With the proliferation of digital educational resources comes the requirement to provide educators with tools for collaboration, customization, and reuse of these resources. We describe our repository software built upon the Drupal content management system that allows educators to form curriculum-based online communities to create, share, and adapt educational resources and curricula. Teachers can create curricula from scratch, add to existing curricula, or clone and customize curricula to fit their teaching environment. We also show how our system can help educators see the usage contexts of their materials. The software is demonstrated in the STEMRobotics middle and high school robotics curricula repository and the MSScience middle school standards-based curriculum collaboration site.

Networks are an effective abstraction for representing real systems. Consequently, network science is increasingly used in academia and industry to solve problems in many fields. Computations that determine structure properties and dynamical behaviors of networks are useful because they give insights into the characteristics of real systems. We introduce a newly built and deployed cyberinfrastructure for network science (CINET) that performs such computations, with the following features: (i) it offers realistic networks from the literature and various random and deterministic network generators; (ii) it provides many algorithmic modules and measures to study and characterize networks; (iii) it is designed for efficient execution of complex algorithms on distributed high performance computers so that they scale to large networks; and (iv) it is hosted with web interfaces so that those without direct access to high performance computing resources and those who are not computing experts can still reap the system benefits. It is a combination of application design and cyberinfrastructure that makes these features possible. To our knowledge, these capabilities collectively make CINET novel. We describe the system and illustrative use cases, with a focus on the CINET user.

Fingerprints continue to serve as a reliable trait for human identification. Feature-based matching techniques, such as those used by Automated Fingerprint Identification Systems (AFIS), have demonstrated remarkable success in minutiae-based matching from good quality prints with relatively large extent. As the image quality degrades and acquired fingerprint area decreases, however, the number of reliable minutiae that can be automatically detected decreases, causing match performance to suffer. This paper presents a novel approach to improving the precision of features that can be extracted from fingerprint images. This is accomplished through improved minutia localization and quality assessment routines that are inspired in part by human visual perception. Initial results have shown an improvement in minutia accuracy for 88.2% of fingerprint minutia sets after applying the proposed localization method. An increase in average quality of true minutiae was found for 98.6% of the fingerprint images when using the proposed quality assessment. The results were obtained using a database of 516 fingerprints with ground truth minutiae.

Search engines are mostly targeted toward the generic user but educators have specific information needs with specialized information-seeking behaviors. When designing a course, educators may create a syllabus, lecture slides, and use tools as lecture aides. These resources are scattered across a large number of websites and require time and effort in finding them. To design a useful system for educators we must understand the problems they face when they seek, use, and re-use online resources. With that in mind, the Ensemble team conducted focus groups to identify current practices and problems in locating online resources for educational purposes. The data provides better understanding of the information-seeking process of educators that can lead to better educational resource sites.

No abstract available.

Researchers have been studying and developing teaching materials for information retrieval (IR), such as [3]. Toolkits also have been built that provide hands-on experience to students. For example, IR-Toolbox [4] is an effort to close the gap between the students' understanding of IR concepts and real-life indexing and search systems. Such tools might be good for helping students in non-technical areas such as in the Library and Information Science field to develop their conceptual model of search engines. However, they do not cover emerging topics and skills, such as content-based image retrieval (CBIR) and fusion search. Although there is open source software (such as those in http://www.searchtools.com/tools/tools-opensource.html) that can be used to teach basic and advanced IR topics, they require a student to have high-level technical knowledge and to spend a long time to gain a practical understanding of these topics.

A syllabus defines the contents of a course, as well as other information such as resources and assignments. In this paper, we report on our work towards creating a syllabus repository of Computer Science courses across universities in the USA. We present some statistics from our initial collection of 8000+ syllabi. We show a syllabus creator that is integrated with Moodle [5] moodle_2006_moodle, an open-source course management system, which allows for the creation of a syllabus for a particular course. Among other information, it includes knowledge units from the Computing Curricula 2001 body of knowledge. The goal of the syllabus repository is to provide added value to the Computer Science Education community, and we present some such offerings. We conclude by presenting our future plans for the syllabus repository. These include using automated techniques to collect and classify syllabi, providing recommendations to instructors when creating a syllabus, and allowing the community to share their syllabi automatically. The syllabus collection will be part of the Computing and Information Technology Interactive Digital Educational Library (CITIDEL), a collection of the National Science Digital Library (NSDL).

In the subsection ''Initial probabilities'' of the Results section, the first sentence was incorrect and should say: These probabilities, denoted by P 0 (RET) and P 0 (LOSS) 1 P 0 (RET), specify the probabilities of RET and LOSS for the pterminal marker on a chromosome.

Support Vector Machines are used to combine the outputs of multiple entity extractors, thus creating a composite entity extraction system. The composite system has a significantly higher f-measure than any of the component systems. Compared to a standard voting technique for combining the results of multiple entity extractors, the SVM approach produces comparable precision and recall statistics but tends to utilize fewer of the component entity extractors, thus providing superior computational efficiency, which is critical in practical applications. In this paper, we present our experimental results of comparing a standard voting technique with SVM that each aggregate four entity extractors. We also describe our future plans of integrating agent-based technology into our experimental testbed where we examine the evolution of composite techniques as part of the analysis stream. Given that much of the improvement comes from tuning the algorithms to the data stream with a human-in-the-loop, we are considering the merits of employing cognitive agents that are strategically embedded in the workflow for processing data. As we tune the algorithms for better performance on the data streams, we envision agents learning the patterns of data streams and apply the appropriate tuning to ensure optimality.

This poster describes the publication of educational resources in ACM's online Journal for Educational Resources in Computing (JERIC) and the US National Science Foundation's National STEM Digital Library (NSDL).

The GetSmart system was built to support theoretically sound learning processes in a digital library environment by integrating course management, digital library, and concept mapping components to support a constructivist, six-step, information search process. In the fall of 2002 more than 100 students created 1400 concept maps as part of selected computing classes offered at the University of Arizona and Virginia Tech. Those students conducted searches, obtained course information, created concept maps, collaborated in acquiring knowledge, and presented their knowledge representations. This article connects the design elements of the GetSmart system to targeted concept-map-based learning processes, describes our system and research testbed, and analyzes our system usage logs. Results suggest that students did in fact use the tools in an integrated fashion, combining knowledge representation and search activities. After concept mapping was included in the curriculum, we observed improvement in students' online quiz scores. Further, we observed that students in groups collaboratively constructed concept maps with multiple group members viewing and updating map details.

Loss of heterozygosity (LOH) of chromosomal regions bearing tumor suppressors is a key event in the evolution of epithelial and mesenchymal tumors. Identification of these regions usually relies on genotyping tumor and counterpart normal DNA and noting regions where heterozygous alleles in the normal DNA become homozygous in the tumor. However, paired normal samples for tumors and cell lines are often not available. With the advent of oligonucleotide arrays that simultaneously assay thousands of single-nucleotide polymorphism (SNP) markers, genotyping can now be done at high enough resolution to allow identification of LOH events by the absence of heterozygous loci, without comparison to normal controls. Here we describe a hidden Markov model-based method to identify LOH from unpaired tumor samples, taking into account SNP intermarker distances, SNP-specific heterozygosity rates, and the haplotype structure of the human genome. When we applied the method to data genotyped on 100 K arrays, we correctly identified 99% of SNP markers as either retention or loss. We also correctly identified 81% of the regions of LOH, including 98% of regions greater than 3 megabases. By integrating copy number analysis into the method, we were able to distinguish LOH from allelic imbalance. Application of this method to data from a set of prostate samples without paired normals identified known regions of prevalent LOH. We have developed a method for analyzing high-density oligonucleotide SNP array data to accurately identify of regions of LOH and retention in tumors without the need for paired normal samples.

In this paper, we present Stepping Stones and Pathways (SSP), an alternative model of building and presenting answers for the cases when queries on document collections cannot be answered just by a ranked list. Stepping Stones can handle questions like: What is the relation of topics X and Y? SSP addresses when the contents of a small set of related documents is needed as an answer rather than a single document, or when query splitting is required to satisfactorily explore a document space. Query results are networks of document groups representing topics, each group relating to and connecting (by documents) to other groups in the network. Thus, a network answers the user's information need. We devise new and more effective representations and techniques to visualize such answers, and to involve users as part of the answer-finding process. In order to verify the validity of our approach, and since the questions we aim to answer involve multiple topics, we performed a study involving a custom built broad collection of operating systems research papers, and evaluated the results with interested computer science students, using multiple measures.

In this paper, we propose a novel framework using Genetic Programming to combine image database descriptors for content-based image retrieval (CBIR). Our framework is validated through several experiments involving two image databases and specific domains, where the images are retrieved based on the shape of their objects.

This paper shows how citation-based information and structural content (e.g., title, abstract) can be combined to improve classification of text documents into predefined categories. We evaluate different measures of similarity -- five derived from the citation information of the collection, and three derived from the structural content -- and determine how they can be fused to improve classification effectiveness. To discover the best fusion framework, we apply Genetic Programming (GP) techniques. Our experiments with the ACM Computing Classification Scheme, using documents from the ACM Digital Library, indicate that GP can discover similarity functions superior to those based solely on a single type of evidence. Effectiveness of the similarity functions discovered through simple majority voting is better than that of content-based as well as combination-based Support Vector Machine classifiers. Experiments also were conducted to compare the performance between GP techniques and other fusion techniques such as Genetic Algorithms (GA) and linear fusion. Empirical results show that GP was able to discover better similarity functions than GA or other fusion techniques.

In this paper we use a Unified Relationship Matrix (URM) to represent a set of heterogeneous data objects (e.g., web pages, queries) and their interrelationships (e.g., hyperlinks, user click-through sequences). We claim that iterative computations over the URM can help overcome the data sparseness problem and detect latent relationships among heterogeneous data objects, thus, can improve the quality of information applications that require com- bination of information from heterogeneous sources. To support our claim, we present a unified similarity-calculating algorithm, SimFusion. By iteratively computing over the URM, SimFusion can effectively integrate relationships from heterogeneous sources when measuring the similarity of two data objects. Experiments based on a web search engine query log and a web page collection demonstrate that SimFusion can improve similarity measurement of web objects over both traditional content based algorithms and the cutting edge SimRank algorithm.

This paper shows how different measures of similarity derived from the citation information and the structural content (e.g., title, abstract) of the collection can be fused to improve classification effectiveness. To discover the best fusion framework, we apply Genetic Programming (GP) techniques. Our experiments with the ACM Computing Classification Scheme, using documents from the ACM Digital Library, indicate that GP can discover similarity functions superior to those based solely on a single type of evidence. Effectiveness of the similarity functions discovered through simple majority voting is better than that of content-based as well as combination-based Support Vector Machine classifiers. Experiments also were conducted to compare the performance between GP techniques and other fusion techniques such as Genetic Algorithms (GA) and linear fusion. Empirical results show that GP was able to discover better similarity functions than other fusion techniques.

We introduce the Multiple Relationship Similarity Spreading Algorithm (MRSSA) to enhance IR effectiveness. This method has similarity computed in an iterative "spreading" fashion for multiple object types, combining both inter- and intra-object relationships. We demonstrate the value of this approach in the context of the WWW, where the key objects are web pages and queries, Relationships considered are derived from hyperlinks (in- and out-links) and click-through logs.

This paper discusses how citation-based information and structural content (e.g., title, abstract) can be combined to improve classification of text documents into predefined categories. We evaluate different measures of similarity derived from the citation structure and the structural content of the collection, and determine how they can be fused to improve classification effectiveness. To discover the best fusion framework, we apply Genetic Programming (GP) techniques. Our empirical experiments using documents from the ACM Digital Library and the ACM Computing Classification System show that we can discover similarity functions that work better than using evidence in isolation and whose combined performance through a simple majority voting is comparable to that of Support Vector Machine classifiers.

This poster describes the publication of educational resources in ACM's online Journal for Educational Resources in Computing (JERIC) and the US National Science Foundation's National STEM Digital Library (NSDL).

This demonstration will illustrate the use of existing and proposed tools in CITIDEL for finding existing resources to integrate into a course and for developing well structured descriptions of class and laboratory activities for sharing with the computing community.

Both ranking functions and user queries are very important factors affecting a search engine's performance. Prior research has looked at how to improve ad-hoc retrieval performance for existing queries while tuning the ranking function, or modify and expand user queries using a fixed ranking scheme using blind feedback. However, almost no research has looked at how to combine ranking function tuning and blind feedback together to improve ad-hoc retrieval performance. In this paper, we look at the performance improvement for ad-hoc retrieval from a more integrated point of view by combining the merits of both techniques. In particular, we argue that the ranking function should be tuned first, using user-provided queries, before applying the blind feedback technique. The intuition is that highly-tuned ranking offers more high quality documents at the top of the hit list, thus offers a stronger baseline for blind feedback. We verify this integrated model in a large scale heterogeneous collection and the experimental results show that combining ranking function tuning and blind feedback can improve search performance by almost 30% over the baseline Okapi system.

Web link analysis has proven to be a significant enhancement for quality based web search. Most existing links can be classified into two categories: intra-type links (e.g., web hyperlinks), which represent the relationship of data objects within a homogeneous data type (web pages), and inter-type links (e.g., user browsing log) which represent the relationship of data objects across different data types (users and web pages). Unfortunately, most link analysis research only considers one type of link. In this paper, we propose a unified link analysis framework, called "link fusion", which considers both the inter- and intra- type link structure among multiple-type inter-related data objects and brings order to objects in each data type at the same time. The PageRank and HITS algorithms are shown to be special cases of our unified link analysis framework. Experiments on an instantiation of the framework that makes use of the user data and web pages extracted from a proxy log show that our proposed algorithm could improve the search effectiveness over the HITS and DirectHit algorithms by 24.6% and 38.2% respectively.

Abstract Genetic‐based evolutionary learning algorithms, such as genetic algorithms (GAs) and genetic programming (GP), have been applied to information retrieval (IR) since the 1980s. Recently, GP has been applied to a new IR task—discovery of ranking functions for Web search—and has achieved very promising results. However, in our prior research, only one fitness function has been used for GP‐based learning. It is unclear how other fitness functions may affect ranking function discovery for Web search, especially since it is well known that choosing a proper fitness function is very important for the effectiveness and efficiency of evolutionary algorithms. In this article, we report our experience in contrasting different fitness function designs on GP‐based learning using a very large Web corpus. Our results indicate that the design of fitness functions is instrumental in performance improvement. We also give recommendations on the design of fitness functions for genetic‐based information retrieval experiments.

Digital libraries (DLs) are complex information systems and therefore demand formal foundations lest development efforts diverge and interoperability suffers. In this article, we propose the fundamental abstractions of Streams, Structures, Spaces, Scenarios, and Societies (5S), which allow us to define digital libraries rigorously and usefully. Streams are sequences of arbitrary items used to describe both static and dynamic (e.g., video) content. Structures can be viewed as labeled directed graphs, which impose organization. Spaces are sets with operations on those sets that obey certain constraints. Scenarios consist of sequences of events or actions that modify states of a computation in order to accomplish a functional requirement. Societies are sets of entities and activities and the relationships among them. Together these abstractions provide a formal foundation to define, relate, and unify concepts---among others, of digital objects, metadata, collections, and services---required to formalize and elucidate "digital libraries". The applicability, versatility, and unifying power of the 5S model are demonstrated through its use in three distinct applications: building and interpretation of a DL taxonomy, informal and formal analysis of case studies of digital libraries (NDLTD and OAI), and utilization as a formal basis for a DL description language.

This demonstration allows conference attendees to gain experience in using the facilities of the NSF National Science Technology Engineering and Mathematics Education Digital Library (NSDL) to prepare and present College or university level classes.

The NSDL consists of collections of materials accessible online, as well as many tools and services intended to enhance learning. Tools and services will support students and as well as faculty. This session will provide a status report on the state of the NSDL, its content, tools, and services. The session includes specific presentations of several services included in the NSDL. Session attendees discuss current and proposed initiatives, and offer suggestions about needs of highest priority.

The Web contains a huge volume of information, almost all unstructured and, therefore, difficult to manage. In Digital Libraries, however, information is explicitly organized, described, and managed. In this paper, we propose an architecture that allows the construction of digital libraries from the Web, using standard protocols and archival technologies, and incorporating powerful digital library and data extraction tools, thus benefiting from the breadth of the Web contents, but supporting services and organization available in digital libraries. The proposed architecture was applied to the Networked Digital Library of Theses and Dissertations, providing an important first step toward rapid construction of large DLs from the Web, as well as a large-scale solution for interoperability between independent digital libraries.

In this demonstration, we present interoperable and personalized search services for the Networked Digital Library of Theses and Dissertations (NDLTD). Using standard protocols and software, including those specified by the Open Archives Initiative (OAI), distributed sites can share metadata easily. On top of these harvesting protocols, we implement a union collection of theses managed by the MARIAN digital library system. Our demonstration covers aspects of NDLTD, OAI, and MARIAN.

No abstract available.

No abstract available.

This panel will discuss a recommendation for curricula guidelines for courses about multimedia. Based on conference input, the guidelines will be modified for review prior to future publication by ACM (with other groups). The recommendation will be one of the main results of a June 2000 Dagstuhl workshop led by 4 of the panelists. Dr. Heller, the 5th panelist, also attended that workshop, and is co-PI with the panel moderator on a directly related NSF project, Curriculum Resources in Interactive Multimedia, CRIM, see http://ei.cs.vt.edu/~crim. Each of the panelists has taught courses about multimedia. Each is interested in developing tools/demonstrations/resources to help in those course. Each has a particular area of interest in the multimedia field. Together they constitute a representative group among those who will contribute to and benefit from curricula resources in multimedia.

The CRIM project addresses the need for curricular guidelines and educational resources for the Interactive Multimedia area. A digital library / repository allows educators to submit knowledge modules that will be reviewed and made available for use by teachers or students. Recommendations are given for courses and topics, and a process is outlined to reach consensus and improve education. This efforts is connected with the Computer Science Teaching Center, http://www.cstc.org/.

No abstract available.

The QUIZIT system has been developed to support adaptive as well as standard testing, along with automatic grading and record keeping, using the WWW as a delivery vehicle. In connection with our Interactive Learning with a Digital Library in Computer Science project, we have applied QUIZIT to help with a master's level course as well as a large freshman level service course. Preliminary testing has led to various improvements, and it appears that QUIZIT will be well received by both instructors and students. It should port easily to a variety of environments and so could be widely used at other institutions as well.

World-Wide Web proxy servers that cache documents can potentially reduce three quantities: the number of requests that reach popular servers, the volume of network traffic resulting from document requests, and the latency that an end-user experiences in retrieving a document. This paper examines the first two using the measures of cache hit rate and weighted hit rate (or fraction of client-requested bytes returned by the proxy). A client request for an uncached document may cause the removal of one or more cached documents. Variable document sizes and types allow a rich variety of policies to select a document for removal, in contrast to policies for CPU caches or demand paging, that manage homogeneous objects. We present a taxonomy of removal policies. Through trace-driven simulation, we determine the maximum possible hit rate and weighted hit rate that a cache could ever achieve, and the removal policy that maximizes hit rate and weighted hit rate. The experiments use five traces of 37 to 185 days of client URL requests. Surprisingly, the criteria used by several proxy-server removal policies (LRU, Hyper-G, and a proposal by Pitkow and Recker) are among the worst performing criteria in our simulation; instead, replacing documents based on size maximizes hit rate in each of the studied workloads.

No abstract available.

A digital library of computer science literature, Envision provides powerful information visualization by displaying search results as a matrix of icons, with layout semantics under user control.Envision's Graphic View interacts with an Item Summary Window giving users access to bibliographic information, and XMosaic provides access to complete bibliographic information, abstracts, and full content.While many visualization interfaces for information retrieval systems depict ranked query-document similarity, Envision graphically presents a variety of document characteristics and supports an extensive range of user tasks.Formative usability evaluation results show great user satisfaction with Envision's style of presentation and the document characteristics visualized.

Much research in informationretrieval haafocused more on matching results to queries than on browsing those restdts.After briefly exploring browsing in physical and electronic libraries, we introduce SortTables, a new system that focuses on support for browsing.We explore the evolution of the system in light of early implementation experience and formative evaluation of the interface.Fh-mlly, we briefly review related work, and discuss future directions.

We describe how to investigate collections of trace data representing network delivery of multimedia information with CHITRA95, a tool that allows a user to visualize, query, statistically analyze and test, transform, and model collections of trace data. CHITRA95 is applied to characterize World Wide Web (WWW) traffic from three workloads: students in a classroom of network-connected workstations, graduate students browsing the Web, undergraduates browsing educational and other materials, as well as traffic on a courseware repository server. We explore the inter-access time of files on a server (i.e., recency), the hit rate from a proxy server cache, and the distributions of file sizes and media types requested. The traffic study also yields statistics on the effectiveness of caching to improve transfer rates. In contrast to past WWW traffic studies, we analyze client as well as server traffic; we compare three workloads rather than drawing conclusions from one workload; and we analyze tcpdump logs to calculate the performance improvement in throughput that an end user sees due to caching.

No abstract available.

This paper describes development of an Display Measurement and Analysis System (DMAS).An object data model for DMAS has been specified, and a prototype database subsystem has been implemented using the ObjectStore ODBMS, accessed from a Windows graphical user interface.The DMAS database was designed to be fully object-oriented to support complex data modelling capabilities and easy extension.Research endeavors include exploring ODBMS in this unconventional engineering application domain and development of an object data model and a testbed for display measurement systems.

No abstract available.

No abstract available.

No abstract available.

Since 1986 we have investigated the problems and possibilities of applying modern information retrieval methods to large online public access library catalogs (OPACs). In the Retrieval Experiment—Virginia Tech OnLine Catalog (REVTOLC) study we carried out a large pilot test in 1987 and a larger, controlled investigation in 1990, with 216 users and roughly 500,000 MARC records. Results indicated that a forms-based interface coupled with vector and relevance feedback retrieval methods would be well received. Recent efforts developing the Multiple Access and Retrieval of Information with Annotations (MARIAN) system have involved used of a specially developed object-oriented DBMS, construction of a client running under NeXTSTEP, programming of a distributed server with a thread assigned to each user session to increase concurrency on a small network of NeXTs, refinement of algorithms to use objects and stopping rules for greater efficiency, usability testing and iterative interface refinement.

Project Envision aims to build a “user-centered database from the computer science literature,” initially using the publications of the Association for Computing Machinery (ACM). Accordingly, we have interviewed potential users, as well as experts in library, information, and computer science—to understand their needs, to become aware of their perception of existing information systems, and to collect their recommendations. Design and formative usability evaluation of our interface have been based on those interviews, leading to innovative query formulation and search results screens that work well according to our usability testing. Our development of the Envision database, system software, and protocol for client-server communication builds upon work to identify and represent “objects” that will facilitate reuse and high-level communication of information from author to reader (user). All these efforts are leading not only to a usable prototype digital library but also to a set of nine principles for digital libraries, which we have tried to follow, covering issues of representation, architecture, and interfacing. © 1993 John Wiley & Sons, Inc.

Our previous research on one-probe access to large collections of data indexed by alphanumeric keys has produced the first practical minimal perfect hash functions for this problem. Here, a new algorithm is described for quickly finding minimal perfect hash functions whose specification space is very close to the theoretical lower bound, i.e., around 2 bits per key. The various stages of processing are detailed, along with analytical and empirical results, including timing for a set of over 3.8 million keys that was processed on a NeXTstation in about 6 hours.

Rapid access to information is essential for a wide variety of retrieval systems and applications.Hashing has long been used when the fastest possible direct search is desired, but is generally not appropriate when sequential or range searches are also required.This paper describes a hashing method, developed for collections that are relatively static, that supports both direct and sequential access.The algorithms described give hash functions that are optimal in terms of time and hash table space utilization, and that preserve any a priori ordering desired, Furthermore, the resulting order-preserving minimal perfect hash functions (OPMPHFS) can be found using time and space that are linear in the number of keys involve~this is close to optimal,

Rapid access to information is essential for a wide variety of retrieval systems and applications. Hashing has long been used when the fastest possible direct search is desired, but is generally not appropriate when sequential or range searches are also required. This paper describes a hashing method, developed for collections that are relatively static, that supports both direct and sequential access. Indeed, the algorithm described gives hash functions that are optimal in terms of time and hash table space utilization, and that preserve any a priori ordering desired. Furthermore, the resulting order preserving minimal perfect hash functions (OPMPHFs) can be found using space and time that is on average linear in the number of keys involved.

Since there are many similarities between the fields of information retrieval and artificial intelligence, it may be appropriate to adapt methods developed in one field to applications usually thought of as belonging to the other. While recent attention in this regard has focused on the use of artificial intelligence methods in the information sciences, this project demonstrates the utility of information retrieval techniques in the expert systems area of artificial intelligence. In particular, it addresses the difficult task of building an expert system for the social sciences. FOCES (FOster Care Expert System) is a prototype assistant to social workers involved in selecting foster homes for children who are in need of placement. The design of FOCES called for use of GUESS (General-pUrpose Expert System Shell) and tailored Prolog routines for extended Boolean matching and vector correlation. Evaluation of the implemented system has shown that FOCES can perform its assigned task as well as trained social workers. © 1990 John Wiley & Sons, Inc.

With the advent of the information society, many researchers are turning to artificial intelligence techniques to provide effective retrieval over large bodies of textual information. In the CODER system, the mission of which is to provide an environment for experiments in applying AI to information retrieval, a factual representation language (FRL) serves as a tool for knowledge engineering and experimentation. the FRL is a hybrid AI language supporting strong typing for attribute values, a frame system, and Prolog-like relational structures. Inheritance is enforced throughout, and the semantics of type subsumption and object matching are formally defined. A collection of type and object managers called the knowledge administration complex implements this common language for storing knowledge and communicating it within the system. Storage of large numbers of complete knowledge objects (statements in the language) is supported by a system of external knowledge bases. Of the three types of knowledge structures in the language, the frame facility has proven most useful in the retrieval domain. This article discusses the frame construct itself, the implementation of the relevant portions of the knowledge administration complex and external knowledge bases, and the use of frames in retrieval research. It closes with a discussion of the utility of the FRL.

This report considers combining information to improve retrieval. The vector space model has been extended so different classes of data are associated with distinct concept types and their respective subvectors. Two collections with multiple concept types are described, ISI-1460 and CACM-3204. Experiments indicate that regression methods can help predict relevance, given query-document similarity values for each concept type. After sampling and transformation of data, the coefficient of determination for the best model was .48 (.66) for ISI (CACM). Average precision for the two collections was 11% (31%) better for probabilistic feedback with all types versus with terms only. These findings may be of particular interest to designers of document retrieval or hypertext systems since the role of links is shown to be especially beneficial.

Experimental information retrieval (IR) systems, some dating back to the sixties, have demonstrated the viability of fully automatic document storage and retrieval methodologies with small to medium size bibliographic collections [72]. Many of these experimental systems utilize the vector space model in which each important term (such as a word stem) identifies a different dimension in a space, so that matrix methods and vector operations can be defined on queries and documents. Statistical techniques have been very effective, and probabilistic enhancements have given additional improvements [84]. However, the basic vector space model is oriented towards recording the essential information in the text of a title/abstract combination rather than describing more complex document structures. It is necessary to extend the model in order to handle composite documents.

Abstract Automatic feedback methods may be used in online information retrieval to generate improved query statements based on information contained in previously retrieved documents. In this study automatic relevance feedback techniques are applied to Boolean query statements. The feedback operations are carried out using both the conventional Boolean logic, as well as an extended logic producing improved retrieval effectiveness. Experimental output is included to evaluate the automatic feedback operations.

Modern information retrieval systems are designed to supply relevant information in response to requests received from the user population. In most retrieval environments the search requests consist of keywords, or index terms, interrelated by appropriate Boolean operators. Since it is difficult for untrained users to generate effective Boolean search requests, trained search intermediaries are normally used to translate original statements of user need into useful Boolean search formulations. Methods are introduced in this study which reduce the role of the search intermediaries by making it possible to generate Boolean search formulations completely automatically from natural language statements provided by the system patrons. Frequency considerations are used automatically to generate appropriate term combinations as well as Boolean connectives relating the terms. Methods are covered to produce automatic query formulations both in a standard Boolean logic system, as well as in an extended Boolean system in which the strict interpretation of the connectives is relaxed. Experimental results are supplied to evaluate the effectiveness of the automatic query formulation process, and methods are described for applying the automatic query formulation process in practice.