Christine Alvarado

The perfect search engine is not enough: a study of orienteering behavior in directed search

This paper presents a modified diary study that investigated how people performed personally motivated searches in their email, in their files, and on the Web. Although earlier studies of directed search focused on keyword search, most of the search behavior we observed did not involve keyword search. Instead of jumping directly to their information target using keywords, our participants navigated to their target with small, local steps using their contextual knowledge as a guide, even when they knew exactly what they were looking for in advance. This stepping behavior was especially common for participants with unstructured information organization. The observed advantages of searching by taking small steps include that it allowed users to specify less of their information need and provided a context in which to understand their results. We discuss the implications of such advantages for the design of personal information management tools.

Women in CS: an evaluation of three promising practices

Historically, Harvey Mudd College (HMC) has had very little success attracting women to the study of computer science: women have chosen CS less than any other field of study. In 2006 HMC began three practices in order to increase the number of women studying and majoring in CS; these practices have now been in place for 3 years. With this paper we describe these practices and present a thorough evaluation of the quantitative and qualitative differences that have accompanied them. In sum, these efforts have rebalanced our department by significantly increasing womens participation in our computer science program.

Multi-domain sketch understanding

People use sketches to express and record their ideas in many domains, including mechanical engineering, software design, and information architecture. In recent years there has been an increasing interest in sketch-based user interfaces, but the problem of robust free-sketch recognition remains largely unsolved. Current computer sketch recognition systems are difficult to construct, and either are fragile or accomplish robustness by severely limiting the designers drawing freedom. This work explores the challenges of multi-domain sketch recognition. We present a general framework and implemented system, called SketchREAD , for diagrammatic sketch recognition. Our system can be applied to a variety of domains by providing structural descriptions of the shapes in the domain. Robustness to the ambiguity and uncertainty inherent in complex, freely-drawn sketches is achieved through the use of context. Our approach uses context to guide the search for possible interpretations and uses a novel form of dynamically constructed Bayesian networks to evaluate these interpretations. This process allows the system to recover from low-level recognition errors (e.g., a line misclassified as an arc) that would otherwise result in domain level recognition errors. We evaluated SketchREAD on real sketches in two domains—family trees and circuit diagrams—and found that in both domains the use of context to reclassify low-level shapes significantly reduced recognition error over a baseline system that did not reinterpret low-level classifications. We discuss remaining challenges for multi-domain sketch recognition revealed by our evaluation. Finally, we explore the systems potential role in sketch-based user interfaces from a human computer interaction perspective. (Copies available exclusively from MIT Libraries, Rm. 14-0551, Cambridge, MA 02139-4307. Ph. 617-253-5668; Fax 617-253-1690.)

Technical Section: The effect of task on classification accuracy: Using gesture recognition techniques in free-sketch recognition

Generating, grouping, and labeling free-sketch data is a difficult and time-consuming task for both user study participants and researchers. To simplify this process for both parties, we would like to have users draw isolated shapes instead of complete sketches that must be hand-labeled and grouped, and then use this data to train our free-sketch symbol recognizer. However, it is an open question whether shapes drawn in isolation accurately reflect the way users draw shapes in a complete diagram. To answer this question, we present a systematic exploration of the effect of task on recognition accuracy using three different recognizers. Our study examines how task affects accuracy in the context of user-independent, user semi-dependent and user-dependent training data. We find that as the amount of user-specific training data increases, the effect of task on recognition accuracy also increases. We also show that the best overall recognition results are obtained by using user semi-dependent, task-specific training data. These results hold across three different domains: circuit diagrams, entity relationship diagrams and process diagrams. Finally, we introduce a variant of a popular and simple gesture recognition algorithm that recognizes freely drawn shapes as well as a highly accurate but more complex recognizer designed explicitly for free-sketch recognition.

A pen-based tool for efficient labeling of 2D sketches

High quality labeled data is essential for developing and evaluating sketch recognition algorithms. Unfortunately, labeling freely-drawn sketches is time-consuming and difficult, if not impossible, using current technologies. These difficulties and the resulting lack of labeled data fundamentally limit the development of recognition algorithms. We present an intuitive, direct manipulation pen-based application for labeling sketch data in any two-dimensional domain. Our labeling tool supports the three essential sketch recognition labeling tasks: stroke fragmentation, stroke grouping and label application. Our interface integrates standard and novel interaction techniques to make each task efficient and natural. In a user study, all users felt that labeling data with our tool was quick and efficient.

CS-1 for scientists

Students in science and engineering are poorly served by most general-purpose CS-1 courses, which rarely discuss scientific problems or applications. At the same time, fewer and fewer computer science students are exposed to scientific ideas or thinking in any of their introductory courses. This divide hurts both sides: scientists and engineers must struggle later in their careers to pick up the computing skills and mindset they need to cope with increasingly computational disciplines, while CS graduates lack the background knowledge needed to work in “relevant” domains ranging from health care to climate modeling. The aim of this panel is to explore what each community thinks it and its students need, and to discuss what is being done to meet those needs at leading institutions. Topics will include:

The effect of task on classification accuracy: using gesture recognition techniques in free-sketch recognition

Generating, grouping, and labeling free-sketch data is a difficult and time-consuming task for both user study participants and researchers. To simplify this process for both parties, we would like to have users draw isolated shapes instead of complete sketches that must be hand-labeled and grouped, and then use this data to train our free-sketch symbol recognizer. However, it is an open question whether shapes draw in isolation accurately reflect the way users draw shapes in a complete diagram. Furthermore, many of the simplest shape recognition algorithms were designed to recognize gestures, and it is not clear that they will generalize to freely-drawn shapes. To answer these questions, we perform experiments using three different recognizers to measure the effect of the data collection task on recognition accuracy. We find that recognizers trained only on isolated shapes can classify freely-sketched shapes as well as the same recognizers trained on free-sketches. We also show that user-specific training examples significantly improve recognition rates. Finally, we introduce a variant of a popular and simple gesture recognition algorithm that recognizes freely-drawn shapes as well as a highly-accurate but more complex recognizer designed explicitly for free-sketch recognition.

What we did: CSEdWeek 2012

During CSEdWeek, computer scientists reach out to their communities to inform public perceptions about what computer science is, the crucial role it plays in our society, the career opportunities it provides, and to launch new efforts to invigorate CS education at all levels. CSEdWeek is celebrated the second week of December to coincide with Grace Hoppers birthday on December 9. Take a look at what some SIGCSE members did to celebrate this year. For more information about CSEdWeek see http://www.csedweek.org/.

NCWIT, from a SIGCSE perspective

The National Center for Women & Information Technology (NCWIT) held its annual summit May 22-24, 2012 in Chicago, IL. As always, the overlap between Summit attendees and SIGCSE members was high, and SIGCSE members got a lot out of their experiences at the Summit. I asked two SIGCSE members to share their experience for those who have never been to an NCWIT Summit.

Multi-domain Hierarchical Free-Sketch Recognition Using Graphical Models

In recent years there has been an increasing interest in sketch-based user interfaces, but the problem of robust free-sketch recognition remains largely unsolved. This chapter presents a graphical-model-based approach to free-sketch recognition that uses context to improve recognition accuracy without placing unnatural constraints on the way the user draws. Our approach uses context to guide the search for possible interpretations and uses a novel form of dynamically constructed Bayesian networks to evaluate these interpretations. An evaluation of this approach on two domains—family trees and circuit diagrams—reveals that in both domains the use of context to reclassify low-level shapes significantly reduces recognition error over a baseline system that does not reinterpret low-level classifications. Finally, we discuss an emerging technique to solve a major remaining challenge for multi-domain sketch recognition revealed by our evaluation: the problem of grouping strokes into individual symbols reliably and efficiently, without placing unnatural constraints on the user’s drawing style.