The TaskTracer system seeks to help multi-tasking users manage the resources that they create and access while carrying out their work activities. It does this by associating with each user-defined activity the set of files, folders, email messages, contacts, and web pages that the user accesses when performing that activity. The initial TaskTracer system relies on the user to notify the system each time the user changes activities. However, this is burdensome, and users often forget to tell TaskTracer what activity they are working on. This paper introduces TaskPredictor, a machine learning system that attempts to predict the user’s current activity. TaskPredictor has two components: one for general desktop activity and another specifically for email. TaskPredictor achieves high prediction precision by combining three techniques: (a) feature selection via mutual information, (b) classification based on a confidence threshold, and (c) a hybrid design in which a Naive Bayes classifier estimates the classification confidence but where the actual classification decision is made by a support vector machine. This paper provides experimental results on data collected from Task-Tracer users.

We report on a field study of the multitasking behavior of computer users focused on the suspension and resumption of tasks. Data was collected with a tool that logged users’ interactions with software applications and their associated windows, as well as incoming instant messaging and email alerts. We describe methods, summarize results, and discuss design guidelines suggested by the findings.

Email-based activity management systems promise to give users better tools for managing increasing volumes of email, by organizing email according to a user’s activities. Current activity management systems do not automatically classify incoming messages by the activity to which they belong, instead relying on simple heuristics (such as message threads), or asking the user to manually classify incoming messages as belonging to an activity. This paper presents several algorithms for automatically recognizing emails as part of an ongoing activity. Our baseline methods are the use of message reply-to threads to determine activity membership and a naïve Bayes classifier. Our SimSubset and SimOverlap algorithms compare the people involved in an activity against the recipients of each incoming message. Our SimContent algorithm uses IRR (a variant of latent semantic indexing) to classify emails into activities using similarity based on message contents. An empirical evaluation shows that each of these methods provide a significant improvement to the baseline methods. In addition, we show that a combined approach that votes the predictions of the individual methods performs better than each individual method alone.

Abstract. Integrated development environments have been designed and engineered to display structural information about the source code of large systems. When a development task lines up with the structure of the system, the tools in these environments do a great job of supporting developers in their work. Unfortunately, many development tasks do not have this characteristic. Instead, they involve changes that are scattered across the source code and various other kinds of artifacts, including bug reports and documentation. Today’s development environments provide little support for working with scattered pieces of a system, and as a result, are not adequately supporting the ways in which developers work on the system. Fortunately, many development tasks do have a structure. This structure emerges from a developer’s actions when changing the system. In this paper, we describe how the structure of many tasks crosscuts system artifacts, and how by capturing that structure, we can make it as easy for developers to work on changes scattered across the system’s structure as it is to work on changes that line up with the system’s structure. 1

Many real-world domains exhibit rich relational structure and stochasticity and motivate the development of models that combine predicate logic with probabilities. These models describe probabilistic influences between attributes of objects that are related to each other through known domain relationships. To keep these models succinct, each such influence is considered independent of others, which is called the assumption of “independence of causal influences ” (ICI). In this paper, we describe a language that consists of quantified conditional influence statements and captures most relational probabilistic models based on directed graphs. The influences due to different statements are combined using a set of combining rules such as Noisy-OR. We motivate and introduce multi-level combining rules, where the lower level rules combine the influences due to different ground instances of the same statement, and the upper level rules combine the influences due to different statements. We present algorithms and empirical results for parameter learning in the presence of such combining rules. Specifically, we derive and implement algorithms based on gradient descent and expectation maximization for different combining rules and evaluate them on synthetic data and on a real-world task. The results demonstrate that the algorithms are able to learn both the conditional probability distributions of the influence statements and the parameters of the combining rules. 1.

Research has shown that computers are notoriously bad at supporting the management of parallel activities and interruptions, and that mobility increases the severity and scope of these problems. This paper presents activity-based computing (ABC) which supplements the prevalent data- and application-oriented computing paradigm with technologies for handling multiple, parallel and mobile work activities. We present the design and implementation of ABC support embedded in the Windows XP operating system. This includes replacing the Windows Taskbar with an Activity Bar, support for handling Windows applications, a zoomable user interface, and support for moving activities across different computers. We report an evaluation of this Windows XP ABC system which is based on a multi-method approach, where perceived ease-of-use and usefulness was evaluated together with rich interview material. This evaluation showed that users found the ABC XP extension easy to use and likely to be useful in their own work.

The ability to detect and differentiate breakpoints during task execution is critical for enabling defer-to-breakpoint policies within interruption management. In this work, we examine the feasibility of building statistical models that can detect and differentiate three granularities (types) of perceptually meaningful breakpoints during task execution, without having to recognize the underlying tasks. We collected ecological samples of task execution data, and asked observers to review the interaction in the collected videos and identify any perceived breakpoints and their type. Statistical methods were applied to learn models that map features of the interaction to each type of breakpoint. Results showed that the models were able to detect and differentiate breakpoints with reasonably high accuracy across tasks. Among many uses, our resulting models can enable interruption management systems to better realize defer-to-breakpoint policies for interactive, free-form tasks.

Helping computer users rapidly locate files in their folder hierarchies has become an important research topic in today’s intelligent user interface design. This paper reports on FolderPredictor, a software system that can reduce the cost of locating files in hierarchical folders. FolderPredictor applies a cost-sensitive prediction algorithm to the user’s previous file access information to predict the next folder that will be accessed. Experimental results show that, on average, FolderPredictor reduces the cost of locating a file by 50%. Another advantage of FolderPredictor is that it does not require users to adapt to a new interface, but rather meshes with the existing interface for opening files on the Windows platform.

Knowledge workers spend the majority of their working hours processing and manipulating information. These users face continual costs as they switch between tasks to retrieve and create information. The TaskTracer project at Oregon State University is investigating the possibilities of a desktop software system that will record in detail how knowledge workers complete tasks, and intelligently leverage that information to increase efficiency and productivity. Our approach combines human-computer interaction and machine learning to assign each observed action (opening a file, saving a file, sending an email, cutting and pasting information, etc.) to a task for which it is likely being performed. In this paper we report on ways we have applied machine learning in this environment and lessons learned so far.

In this electronic age, we are all knowledge workers, tackling on daily basis the information that flows through our email, file systems, web browsers, calendars, and various other desktop applications. Email has come to be the center of desktop activity for many of us: we set up meetings, exchange documents, manage projects, forward interesting links, track requisitions, and perform a whole host of tasks through email. Many of these activities involve information flowing from one action to the next, across applications. They are also often repetitive, structured procedures, involving an ordered set of steps, many but not all of which are amenable to automation. In this paper, we present our approach to learning generalized dataflow procedures on the desktop from demonstrations by the user. Our system LAPDOG runs on the CALO desktop assistant and is capable of acquiring fully or partially automated procedures, spanning multiple applications, from demonstrations where the user may perform some unobservable actions.