Paper Reading #20: iSlideShow: a content-aware slideshow system

Reference Information

Title: iSlideshow: a Content-Aware Slideshow System
Authors: Jiajian Chen, Jun Xiao, Yuli Gao
Conference: IUI '10 Proceedings of the 15th international conference on Intelligent user interfaces

Summary

The authors present an intelligent photo slideshow system that automatically analyzes thematic information about the photo collection and utilizes such information to generate compositions and transitions in two modes: story-telling mode and person-highlighting mode. In the story-telling mode the system groups photos by a theme-based clustering algorithm and multiple photos in each theme cluster are seamlessly tiled on a slide. Multiple tiling layouts are generated for each theme cluster and the slideshow is animated by intra-cluster transitions. In the person-highlighting mode, the system first recognizes faces from photos and creates photo clusters for individuals. It then uses face areas as ROI (Regions of Interests) and creates various content-based transitions to highlight individuals in a cluster. With an emphasis on photo content, our system creates slideshows with more fluid, dynamic and meaningful structure compared to existing systems.

Figure 1.0 iSlideShow Interface

Discussion

Paper Reading 19: From documents to tasks: deriving user tasks from document usage patterns

Reference Information

Title: From Documents to Tasks: Deriving User Tasks from Document Usage Patterns
Authors: Oliver Brdiczka
Conference: IUI '10 Proceedings of the 15th international conference on Intelligent

Summary

A typical knowledge worker is involved in multiple tasks and switches frequently between them every work day. These frequent switches become expensive because each task switch requires some recovery time as well as the reconstitution of task context. First task management support systems have been proposed in recent years in order to assist the user during these switches. However, these systems still need a fairly big amount of investment from the user side in order to either learn to use or train such a system.

In order to reduce the necessary amount of training, this paper proposes a new approach for automatically estimating a user’s tasks from document interactions in an unsupervised manner. While most previous approaches to task detection look at the content of documents or window titles, which might raise confidentiality and privacy issues, our approach only requires document identifiers and the temporal switch history between them as input.

The prototype system monitors a user’s desktop activities and logs documents that have focus on the user’s desktop by attributing a unique identifier to each of these documents. Retrieved documents are filtered by their dwell times and a document similarity matrix is estimated based on document frequencies and switches. A spectral clustering algorithm then groups documents into tasks using the derived similarity matrix. The described prototype system has been evaluated on user data of 29 days from 10 different subjects in a corporation. Obtained results indicate that the approach is better than previous approaches that use content.

(Figure 1: Average precision, recall and F-measure with respect

to the number of user tasks)

Discussion

This paper was pretty technical. The authors, no doubt, comes from a highly technical background and his writing reflects this fact. However, I has taken the information storage and retrieval class last semester and I am quite familiar with the terms - clustering, F-measure, precision, recall, etc. that he uses in the paper. I think this system will be very useful since we switch between numerous tasks multiple times each day. Software like this would significantly improve work efficiently and reduce the downtime caused due to task switching.

Paper Reading 18: Evaluating the design of inclusive interfaces by simulation

Reference Information:

Title: Evaluating the Design of Inclusive Interfaces by Simulation.
Authors: Pradipta Biswas, Peter Robinson.
Conference: IUI '10 Proceedings of the 15th international conference on Intelligent user interfaces.

Summary:

The authors have developed a simulator to help with the design and evaluation of assistive interfaces. The simulator can predict possible interaction patterns when undertaking a task using a variety of input devices, and estimate the time to complete the task in the presence of different disabilities.

They have evaluated the simulator by considering a representative application being used by ablebodied, visually impaired and mobility impaired people. In the study that they conducted, they compared the simulator's predictions of the time it would take for people with impairment for performing various tasks vs. the time it would take for accomplishing these tasks for other users.

The simulator predicted task completion times for all three groups with statistically significant accuracy. The simulator also predicted the effects of different interface designs on task completion time accurately.

(Fig 1.0 Use of the simulator)

Discussion:

After reading this paper, I have an opinion that this is not a paper that presents a research idea or something that's new in the field. Instead, it presents results of the user study that the researchers conducted. They present their findings and this valuable piece of data can be used by other researchers to develop new applications and systems.

Paper Reading #17: Mobia Modeler: easing the creation process of mobile applications for non-technical users

Reference Information:

Title: Mobia Modeler: Easing the Creation Process of Mobile Applications for Non-Technical Users

Authors: Florence Balagtas-Fernandez, Max Tafelmayer, Heinrich Hussmann

Conference: IUI '10 Proceedings of the 15th international conference on Intelligent user interfaces

Summary:

In this paper, the authors talk about the tools that they developed called the Mobia Modeler. This tool is aimed at the users without technical background or programming expertise. It facilitates these users to develop mobile phone apps.

The development of mobile applications has now extended from mobile network providers into the hands of ordinary people as organizations and companies encourage people to come up with their own software masterpieces by opening up SDKs, APIs and other development tools.

However, as of the moment, these APIs and tools are only usable by people with programming skills. There is a scarcity of tools that enable users without programming experience to easily build customized mobile applications.

The paper talks about Platform Independent Model and Platform Specific Model. The users create apps using the graphical user interface and the program translates the graphical model to the code taking into consideration the relationship between different modules.

In this paper, the authors present a tool and framework that would enable non-technical people to create their own domain-specific mobile applications. The tool features a simple user-interface that features configurable components to easily create mobile applications. As a proof of concept, they focus on the creation of applications in the domain of mobile health monitoring.

Discussion:

I don't think this program can be used to develop complex applications, however, it can be very easily used to develop simple apps designed to solve simple tasks. I am currently developing a phone app and the process is indeed time-consuming. I can see how this program will generate a huge inflow of app developers and an exponential increase in the number of apps available in the app market.

Paper Reading #16: Mixture Model based Label Association Techniques for Web Accessibility

Reference Information:

Title: Mixture Model based Label Association Techniques for Web Accessibility

Authors: Muhammad Asiful Islam, Yevgen Borodin, I. V. Ramakrishnan

Conference: UIST '10 Proceedings of the 23nd annual ACM symposium on User interface software and technology

Summary:

The internet used to consist of text-based pages, but today the internet has grown into a massive chunk of data. The information is growing exponentially each day. However, many websites are not designed for blind users. An important aspect of making the Web accessible to blind

users is ensuring that all important web page elements such

as links, clickable buttons, and form fields have explicitly

assigned labels. Properly labeled content is then correctly

read out by screen readers, a dominant assistive technology

used by blind users.

Improperly labeled form

fields can critically impede online transactions such as

shopping, paying bills, etc. with screen readers. Very often

labels are not associated with form fields or are missing

altogether, making form filling a challenge for blind users.

Algorithms for associating a form element with one of several

candidate labels in its vicinity must cope with the variability

of the element’s features including label's location

relative to the element, distance to the element, etc. Probabilistic

models provide a natural machinery to reason with

such uncertainties. In this paper, the authors present a Finite Mixture

Model (FMM) formulation of the label association problem.

The variability of feature values are captured in the FMM

by a mixture of random variables that are drawn from parameterized

distributions. Then, the most likely label to be

paired with a form element is computed by maximizing the

log-likelihood of the feature data using the Expectation

Maximization algorithm.

The FMM approach was also adopted for two related problems: assigning labels from an external

Knowledge Base to form elements that have no candidate

labels in their vicinity and for quickly identifying clickable

elements such as add-to-cart, checkout, etc., used in online

transactions even when these elements do not have textual

captions e.g., image buttons without alternative text. The authors provide

a quantitative evaluation of their techniques, as well as

a user study with two blind subjects who used an aural web

browser implementing the approach.

The results of the experiments showed that after using VSM, 95% of the web elements were correctly labeled. Couple of blind users were asked questions and feedback to rate the system on the Likert scale. Both the users gave a very postive feedback and mentioned that they've always had problems filling out the online forms but this task was made easy after the web elements in the forms were properly labeled.

Discussion:

This idea is one example of how technology can be used to change lives. Technology has tremendous potential to benefit people. The paper does a great job at increasing the accessibility of the web for the blind users.

Overall, the paper was quite technical and presented the reader with complex mathematical formulas and code segments that were not quite comprehensible in the first reading.

Book Reading #41: Why We Make Mistakes

Summary:

Chapter 2:

This chapter sounds like a continuation of the first chapter, however, now he changes the focus from vision to meanings. He claims that meanings matter while details don't, which is why we forget names since names don't have much of a meaning. Then he talks about slip-of-tongue errors and gives an interesting example of the slip-of-tongue error committed by Joe Theismann on national media. The author also talks about why we forge passwords and hiding places. Finally, he talks about why eye witnesses make errors in identifying the suspects.

Chapter 3:

In this chapter, the author talks about how we connect the dots in our memory to form a complete image. In one of the examples he mentions that during a wine tasting event, people rated the more expensive wine to be the better one even though there was the same wine in one of the bottles marked with a cheaper price tag. He also mentions that smell and color play an important role in forming perceptions, for example, pink is a feminine color while black demonstrates power.

Discussion:

The examples make this book very interesting. The experiment in which people were tested if they remembered the exact details of a penny proves that we don't store details in the memory. Also, he mentions that we must come up with passwords quickly else we ourselves end up forgetting it. I think this was pretty interesting. Also, the wine bottle example that he provides in chapter 3, was really fascinating.

Book Reading #39: Why We Make Mistakes

Reference Information
Title: Why We Make Mistakes
Author: Joseph T. Hallinan
Publisher: 2009 Crown Archetype

Summary:

Introduction:

I thought this chapter was very similar to chapter 5 from the book design of everyday things since both these chapters address human errors and their types. He also mentions some interesting facts like most of us show a preference for the number 7. He also talks about how people keep forgetting their passwords. He states the we make mistakes but blame the wrong cause. In most cases, the human is not to blame but the poor design of the machine is.

Chapter 1:

This chapter talks about our perception about our vision. The author mentions that we don't really see as much as we think we see. Many a times we just look, but we don't really see i.e. we get a general image, but we don't focus on details. He also mentions some interesting human tendencies like we are more likely to look at our right first. Author says that we don't find something often, we tend to miss it. To prove this he mentions error rate of radiologists in finding cancerous tumors.

Discussion:

I enjoyed both these chapters. The examples provided by the author are very interesting and sometimes surprising. I think knowing why we make mistakes can help us to avoid mistakes.