The research in this dissertation consists of two parts: An active learning algorithm for hierarchical labels and an embedding-based retrieval algorithm. In the first part, we present a new approach for learning hierarchically decomposable concepts. The approach learns a high-level classifier (e.g., location vs. non-location) by separately…

Bag-of-words is a commonly used text representation method for many text classification applications. However, bag-of-words representation fails to consider the context of the text because it only examines text documents based on the presence of individual words and explores relationships between texts with similar word choices (Bengfort, 2018).…

Due to the large volume and exponential growth of biomedical documents (e.g., books, journal articles), it has become increasingly challenging for biomedical search engines to retrieve relevant documents based on users' search queries. Part of the challenge is the matching mechanism of free-text indexing that performs matching based on…

Ample evidence exists to support the conclusion that enterprise search is failing its users. This failure is costing corporate America billions of dollars every year. Most enterprise search engines are built using web search engines as their foundations. These search engines are optimized for web use and are inadequate when used inside the…

The inverted index is the main data structure used by all the major search engines. Search engines build an inverted index on their collection to speed up query processing. As the size of the web grows, the length of the inverted list structures, which can easily grow to hundreds of MBs or even GBs for common terms (roughly linear in the size of…

Identifying topics of a textual document is useful for many purposes. We can organize the documents by topics in digital libraries. Then, we could browse and search for the documents with specific topics. By examining the topics of a document, we can quickly understand what the document is about. To augment the traditional manual way of topic…

This study is a comparison of the descriptive tagging practices among library, archive, and museum professionals using an inter-indexing consistency approach. The first purpose of this study was to determine the extent of the similarities and differences among professional groups when assigning descriptive tags to a wide variety of objects that…

Graphs are widely used to model structures and relationships of objects in various scientific and commercial fields. Chemical molecules, proteins, malware system-call dependencies and three-dimensional mechanical parts are all modeled as graphs. In this dissertation, we propose to mine and index those graph data to enable fast and scalable search.…

This dissertation focuses on investigating whether users will locate desired images more efficiently and effectively when they are provided with information descriptors from both experts and the general public. This study develops a way to support image finding through a human-computer interface by providing subject headings and social tags about…

We have developed statistical techniques to improve the performance of document image search systems where the intermediate step of OCR based transcription is not used. Previous research in this area has largely focused on challenges pertaining to generation of small lexicons for processing handwritten documents and enhancement of poor quality…

Hierarchical folders have been widely used for managing digital files. A well constructed hierarchical structure can keep files organized. A parent folder can have several subfolders and one subfolder can only reside in one parent folder. Files are stored in folders or subfolders. Files can be found by traversing a given path, going through…

In response to the rapid development of digital libraries over the past decade, researchers have focused on the use of metadata as an effective means to support resource discovery within online repositories. With the increasing involvement of libraries in digitization projects and the growing number of institutional repositories, it is anticipated…

In recent years, tags have become a standard feature on a diverse range of sites on the Web, accompanying blog posts, photos, videos, and online news stories. Tags are descriptive terms attached to Internet resources. Despite the rapid adoption of tagging, how people use tags during the search process is not well understood. There is little…

Researchers that investigate the media's coverage of health have historically relied on keyword searches to retrieve relevant health news coverage, and manual content analysis methods to categorize and score health news text. These methods are problematic. Manual content analysis methods are labor intensive, time consuming, and inherently…

Recognition and separation of sounds played by various instruments is very useful in labeling audio files with semantic information. This is a non-trivial task requiring sound analysis, but the results can aid automatic indexing and browsing music data when searching for melodies played by user specified instruments. Melody match based on pitch…