George W. Furnas

Indexing by Latent Semantic Analysis

A new method for automatic indexing and retrieval is described. The approach is to take advantage of implicit higher-order structure in the association of terms with documents (“semantic structure”) in order to improve the detection of relevant documents on the basis of terms found in queries. The particular technique used is singular-value decomposition, in which a large term by document matrix is decomposed into a set of ca. 100 orthogonal factors from which the original matrix can be approximated by linear combination. Documents are represented by ca. 100 item vectors of factor weights. Queries are represented as pseudo-document vectors formed from weighted combinations of terms, and documents with supra-threshold cosine values are returned. initial tests find this completely automatic method for retrieval to be promising.

Generalized fisheye views

In many contexts, humans often represent their own “neighborhood” in great detail, yet only major landmarks further away. This suggests that such views (“fisheye views”) might be useful for the computer display of large information structures like programs, data bases, online text, etc. This paper explores fisheye views presenting, in turn, naturalistic studies, a general formalism, a specific instantiation, a resulting computer program, example displays and an evaluation.

The vocabulary problem in human-system communication

In almost all computer applications, users must enter correct words for the desired objects or actions. For success without extensive training, or in first-tries for new targets, the system must recognize terms that will be chosen spontaneously. We studied spontaneous word choice for objects in five application-related domains, and found the variability to be surprisingly large. In every case two people favored the same term with probability <0.20. Simulations show how this fundamental property of language limits the success of various design methodologies for vocabulary-driven interaction. For example, the popular approach in which access is via one designers favorite single word will result in 80-90 percent failure rates in many common situations. An optimal strategy, unlimited aliasing, is derived and shown to be capable of several-fold improvements.

Recommending and evaluating choices in a virtual community of use

When making a choice in the absence of decisive first-hand knowledge, choosing as other like-minded, similarly-situated people have successfully chosen in the past is a good strategy — in effect, using other people as filters and guides: filters to strain out potentially bad choices and guides to point out potentially good choices. Current human-computer interfaces largely ignore the power of the social strategy. For most choices within an interface, new users are left to fend for themselves and if necessary, to pursue help outside of the interface. We present a general history-of-use method that automates a social method for informing choice and report on how it fares in the context of a fielded test case: the selection of videos from a large set. The positive results show that communal history-of-use data can serve as a powerful resource for use in interfaces.

Information Retrieval using a Singular Value Decomposition Model of Latent Semantic Structure

In a new method for automatic indexing and retrieval, implicit higher-order structure in the association of terms with documents is modeled to improve estimates of term-document association, and therefore the detection of relevant documents on the basis of terms found in queries. Singular-value decomposition is used to decompose a large term by document matrix into 50 to 150 orthogonal factors from which the original matrix can be approximated by linear combination; both documents and terms are represented as vectors in a 50- to 150- dimensional space. Queries are represented as pseudo-documents vectors formed from weighted combinations of terms, and documents are ordered by their similarity to the query. Initial tests find this automatic method very promising.

Space-scale diagrams: understanding multiscale interfaces

Big information worlds cause big problems for interfaces. There is too much to see. They are hard to navigate. An armada of techniques has been proposed to present the many scales of information needed. Space-scale diagrams provide an analytic framework for much of this work. By representing both a spatial world and its different magnifications explicitly, the diagrams allow the direct visualization and analysis of important scale related issues for interfaces.

Pad++: A Zoomable Graphical Sketchpad For Exploring Alternate Interface Physics

We describe Pad++, a zoomable graphical sketchpad that we are exploring as an alternative to traditional window and icon-based interfaces. We discuss the motivation for Pad++, describe the implementation, and present prototype applications. In addition, we introduce an informational physics strategy for interface design and briefly contrast it with current design strategies. We envision a rich world of dynamic persistent informational entities that operate according to multiple physics specifically designed to provide cognitively facile access and serve as the basis for design of new computationally-based work materials.

Pictures of relevance: a geometric analysis of similarity measures

We want computer systems that can help us assess the similarity or relevance of existing objects (e.g., documents, functions, commands, etc.) to a statement of our current needs (e.g., the query). Towards this end, a variety of similarity measures have been proposed. However, the relationship between a measures formula and its performance is not always obvious. A geometric analysis is advanced and its utility demonstrated through its application to six conventional information retrieval similarity measures and a seventh spreading activation measure. All seven similarity measures work with a representational scheme wherein a query and the database objects are represented as vectors of term weights. A geometric analysis characterizes each similarity measure by the nature of its iso‐similarity contours in an n‐space containing query and object vectors. This analysis reveals important differences among the similarity measures and suggests conditions in which these differences will affect retrieval performance. The cosine coefficient, for example, is shown to be insensitive to between‐document differences in the magnitude of term weights while the inner product measure is sometimes overly affected by such differences. The context‐sensitive spreading activation measure may overcome both of these limitations and deserves further study. The geometric analysis is intended to complement, and perhaps to guide, the empirical analysis of similarity measures.

Multitrees: enriching and reusing hierarchical structure

This paper introduces multitrees, a new type of structure for representing information. Multitrees are a class of directed acyclic graphs (DAGs) with the unusual property that they have large easily identifiable substructures that are trees. These subtrees have a natural semantic interpretation providing alternate hierarchical contexts for information, as well as providing a natural model for hierarchical reuse. The numerous trees found within multitrees also afford familiar, tree-based graphical interactions.

New graphical reasoning models for understanding graphical interfaces

This paper aspires to make three points: (1) that certain graphical interfaces are especially easy to learn and use, (2) that special graphical deduction / computation systems are possible, and (3) that perhaps points (1) and (2) are intimately related, i.e., that graphical interfaces may be especially useful because they engage special human graphical reasoning processes.