School of Information Science faculty are a proactive unit that seeks not simply to document the world, but to change it. Our research helps farmers in Africa access the latest in agricultural research. We work with the Ministry of Education in Italy to improve learning outcomes for school children. We work with incarcerated youth to develop graphic novels that give the authors agency and self-worth.
Research of our school scholars crosses many disciplines. This list provides a sample of the areas the faculty explore.
Metadata is the creation of surrogates for objects, documents, or concepts. The most pragmatic definition is, “everything you need to know about something without actually having that thing.” What does Amazon have to tell you about a book or a rake that will allow you to make a purchasing decision? When you search for Google and get back a list of results, the title and two sentences or so are metadata. Image search? Those thumbnails are metadata. Metadata can describe an item (topic, length, author) or how that item can be used (who has access, what format is it in).
Our researchers explore what are the necessary elements of metadata for a given use (called schema development), and how systems can use metadata (such as metadata standards for library catalogs). The act of creating metadata in libraries is called cataloging. Metadata can be created by humans (such as a library catalog record or an Amazon description) or through software. For example, Google extracts a lot of information about a webpage automatically, including when it was updated, what links to it, how often people select it in search results, whether it is optimized for mobile devices, and so on. The rise of machine learning and artificial intelligence algorithms have turned this into a particularly hot area of research where abstracts and document summaries can be automatically extracted from texts, images, or even sound files.
One of the key goals of metadata is the organization of large collections of things (documents, books, artifacts). When you have to find an item in thousands or millions of documents, can you narrow your search by topic (subject classification) or by its creation date or author? If it is by author, how can you ensure that the author of one book is identical to another (authority files)? How can you ensure that a document is an original (provenance)? In a communications context, how can you organize video libraries of thousands of hours in a way to make them quickly findable? If you seek to share materials, say courses, between organizations, how can you match how you organize them to the other organization? When you seek to analyze a million tweets, how can you group them by sender, time, place, or some other attribute? Information organization is all about making large collections discoverable and useful.
How do people look for information? In the era of fake news, how do people make credibility judgments on stories? This broad area also covers the concept of “information literacy” that seeks to understand how people transform what they read/watch/listen to into knowledge and action. In a library context, this is often referred to as reference, or in the broader journalism world as research. Information seeking looks at people and groups and attempts to determine patterns in how they seek to find answers to their questions.
Information Use and User Experience
Closely related to information seeking is designing and understanding how people interact with systems (digital and otherwise). What makes an effective website? How can technology, such as RFID, change people’s shopping experiences? This area is also part of what would be called Human Centered Computing (HCC) or Human Computer Interaction (HCI). Recent work in this area includes using critical theory to understand how systems and interface shape people. What is a user/system interaction in the area of biohacking where people are implanting chips and sensors into their bodies? How do VR and augmented reality interfaces affect work environments? How do technology developers embed their own values and world views into these systems (technology is not neutral)?
Information Retrieval and Natural Language Processing
Human beings are semantic beings. We communicate meaning embedded in symbols and signs. How can technology be used to extract meaning from these semantic objects? Or, in English, how can we quickly find meanings in millions of documents? When you do a search for “apple,” how does Google know to search for the fruit versus the technology company? When an intelligence analyst is faced with thousands of ISIS tweets in Arabic, how can they quickly find patterns and represent those patterns visually? One area of particular note in this area is the rise of digital humanities, and the analysis of texts with automation.
Related to Natural Language Processing is data science and analysis. This area includes the use of analytics, but it also involves the lifecycle of data. Data has a lifecycle that spans data creation (measurement, observation, reporting, extraction from texts), to data hygiene (ensuring the data is valid and reliable over time and people involved), to data storage (databased design and management), to data analysis, to the ethical use of data and impacts of use. Projects here cover everything from tracking political use of twitter, to processing terabytes of travel data to improve smart cities, to interface design, to here at USC, ensuring data in the data warehouse is accurate and consistent. While parts of this field are very new, parts of this area go back 4,000 years to the creation of archives and ontologies.
Where data science focuses on explicit information in an organization, knowledge management seeks to capture the tacit knowledge within the members of an organization. What we know is not completely represented in documents or systems. Knowledge management seeks to understand what people know and how they use it in context to make decisions. Examples include capturing experiences of senior and retiring engineers at Boeing to support aircraft still in service decades after production has stopped. Knowledge management tends to see an organization (community, business, government) as complex systems that can be understood by seeking regularities the underlying rules that govern work.
Policy areas of intellectual property, privacy, equitable use of community resources, net neutrality, and public support of learning organizations effect every other part of the field. Information policy is studying the difference between what a community can do and what it should do and the process to determine the differences. There I a strong line of research and conversation around professional ethics and determining beneficial impacts of information.
Library and Information Science is an academic and scholarly discipline with strong roots in the social sciences that has strongly aligned professions. Therefore, there is ample research in how information organizations such as libraries are built and maintained. This work includes: large scale strategies for organizational success; how librarians and the libraries they build can increase learning in students; how librarians fit into workplaces such as governments or media companies; and what skills are needed as an information professional. Within this area there is a strong tradition of social justice and a quest for equity of access and opportunity.
Faculty Research and Areas of Interest
Want to see specific area of interests of our faculty? Visit our Faculty Research webpage.