Meeting Summarization

Paper: Improving Meeting Summarization by Focusing on User Needs: A Task-Oriented Evaluation

Written by: Pei-Yun Hsueh and Johanna D. Moore

Comments: Brandon

Summary:

This paper conducted a study on recorded meetings and how fast they could extract the decisions made in the meetings. The goal was to make a good meeting browser so that people could see summaries of the meeting and figure out the decisions that were made in each meeting. There were 35 participants where 20 of them were female and 15 were male. They filled out a questionnaire about their prior experience with computer usage and meetings, and then they had to analyze four meetings and summarize what decisions were made in the meetings. They had to write a brief summary of the decisions in 45 minutes using the browser interface. The meeting browser had audio capability so that every participant could listen to the meeting or select portions to listen to. The browser also had four different summary displays.

1. Baseline (AE-ASR): automatic general purpose extracts with automatic speech recognition
2. AD-ASR: automatic decision focused extracts with automatic speech recognition
3. AD-REF: automatic decision-focused extracts with manual transcription
4. Topline (MD-REF): manual decision focused extracts with manual transcription

The graph above are the results from the analysis of the summaries that each participant created from the different browser options. It shows that on average the participants that were given the decision-focused extractive summaries were able to hit each decision point much better than the general purpose one. They again tested this to see why there was a difference with an analysis of variance. The decision-focused displays seemed much easier to use to the participants and they thought it was less demanding as well.

Having the display option for the user was beneficial to all participants in achieving their task. It was much easier for the participants to go through the decision summaries in order to figure out the decision points then it was to go through the extractive summaries that were usually rather long. Only a few participants actually went through the audio-video recordings that were made available in order to find a few more of the decisions. The conclusion that the paper came to was that they couldn't conclude how efficient their browser was since the baseline decisions were still of high quality. The topline decision-focused participants were able to go through the task much quicker, but since all participants concluded around the same decisions they could not pinpoint the efficiency of the browser. Even when the users encountered 30-40% inconsistency with the data from the actual meeting being translated into the extracted summary they were still able to come up with what the decisions were. So basically their browser worked well for either scenario, but they couldn't tell us how efficient it was.

Discussion:
This paper was semi-interesting and more just confusing. For some people it may be nice to have this to look back on a few things they were uncertain of or allow people that could not make the meeting see what the meeting was about, but I think that with the way businesses are being run that anyone could attend a meeting from wherever they were and this wouldn't be an issue. I think this browser could be improved and made more useful for a business to use if they found a way to highlight the decision points without someone having to go through the meeting and picking the decisions that were made. I wish I understood their methods a little bit better though and hopefully after I read it again for the presentation I can talk more about it.

IW - Intelligent Wheelchair

Paper: Intelligent Wheelchair (IW) Interface using Face and Mouth recognition

Written by: Jin Sun Ju, Yunhee Shin, and Eun Yi Kim

Comments: Ross

Summary:
This paper proposed the idea of an intelligent wheelchair that would recognize face and mouth controls from a severely disabled person in order to navigate the wheelchair. The direction that the IW would go would depend on where the users face was inclined, and the starting/stopping of the IW would depend on the users' mouth. There are also 10 range sensors used for the wheelchair so that it can avoid obstacles in whatever environment it is in. The wheelchair has an embedded computer and sensors that allow the user to control the movement of the chair. The system is also able to figure out intentional behavior (looking to the right to move the chair) versus unintentional behavior (looking at something that could be a potential obstacle).

This system recognizes a users movements and converts it into something that the chair can understand so that it can move accordingly. There are three steps in this process:

1. Detector
2. Recognizer
3. Converter

Once the facial recognition is in place, it is then time to figure out what the mouth is doing. The way the mouth is detected is based on two properties:

1. The mouth is located in the lower region of the face
2. The mouth has high contrast when compared with its' surroundings

There are a few mouth 'templates' that are taken to begin the recognition of what the mouth is doing (telling the chair to go or stop). K-means clustering is used to figure this out, and then the Hamming distance is used to figure out what mouth template will correspond to a given candidate that is using the chair.

The converter portion is used in order to give a certain amount of power to the machine based on the users input that the compute is reading off. The board program will control the speed by regulating the amount of voltage that the chair will receive.



The authors wanted to test their solution against solutions that have already been created. They tested it against the joystick wheelchairs that most people are used to seeing, and they also compared it to the headband method of moving a chair which is a little bit similar to their solution. Their first experiment dealt with the commands while the second experiment compared their method with the other two methods. They used a range of people in order to test this which is shown below.


Their results showed that their system could accurately detect the face and mouth of each user, and also figure out how their head was inclined so that the IW could move in that direction. The calculated how long each action would take and also compared that to the other methods. They used recall and precision in order to show their methods performance.

They came to the conclusion that the joystick method was the fastest because it more comfortable for people to use because they did not really want to use their face to move the chair. After training the individuals on how to use their face it became evident that their method was more accurate.

Discussion:
I think this paper was very interesting and it would be really nice to have for those people who are severely disabled and are unable to use the joystick wheelchairs. There has been some future work already being planned for this wheelchair such as adding more sensors to accurately detect obstacles to avoid injury to the user of the chair. I think that is a great idea because there are multiple things that could be seen as an obstacle to a wheelchair and it is necessary for this chair to help the user avoid them. They also want to make sure that IW can better detect situations so that when a user is talking to someone the IW won't start moving around on them. This would definitely be something useful for people and I hope that they can improve on it so that people are more comfortable with using it.

IKEA Hacking

Paper: Learning from IKEA Hacking: "I'm Not One to Decoupage a Tabletop and Call It a Day."

Written by: Deniela Rosner and Jonathan Bean

Comments: Jill

Summary:

This paper deals with multiple people taking IKEA furniture and turning it into their own creations. Each person enjoys the design process and really gets satisfaction from it. One user said that they really personalize everything because it was a reflection of themselves in their work. One person modified a chair, and another modified cabinets to go into their kitchen where they wanted them.

Many people hack items for their home use, and this touch based on how people go about doing it. There are now websites that explain how to hack certain IKEA items and a community for people who like to express themselves with this type of creativity. The community allows them to connect with other people who are like them so that they can share their creativity with people who will truly appreciate it.

Discussion:

I thought this was extremely interesting because I've put together IKEA furniture before and not once have I not followed the directions. It would be cool to see what kind of things that a person can make with certain IKEA furniture. I'm glad that they tried to differentiate between code/software hacking and the "hacking" of other items. I'm not sure these can really be compared, but at least they made the effort to differentiate them.

Social Computing - Privacy

Article Title: Social Computing Privacy Concerns: Antecedents & Effects

Written by: Oded Nov and Sunil Wattal

Comments: Eric

Summary:
This paper focused on privacy in a social network setting and how much information a person would share about themselves in certain social communities. Privacy concerns are not yet clear when it comes to social communities, and this paper researches two questions:

1. Antecedents of privacy concerns: "Do social norms contribute to individual's privacy concerns in online communities? Does trust in other network participants affect users' privacy concerns?"
2. Impact of privacy concerns: "Do higher levels of privacy concerns in an online community lead to lower levels of information sharing? What role does the position of the user within the network structure play in a user's willingness to share?"

The way they researched these questions was studying the way photos are shared on Flickr. There were 1270 people invited to do a web survey, and they only received 192 valid responses. Below is a picture of the sample questions asked in the survey.

This paper concluded that when people are concerned with privacy then they will not share as much information, but when it comes to a site like Flickr and other social computer sites they are more likely to share more information because they are familiar with the other users.

Discussion:
For this paper I thought the results would be similar to what they were. Things like Facebook where you know the people you are adding as friends, and the fact that you can control your privacy settings makes it easier for someone to share more information. If it's a new site that they aren't used to then they will probably share less. For photos on the other hand I would have thought people would share less or not allow their personal photos to be shown to the public community, but I guess it has the same effect as Facebook would. I think that for the future work it would be beneficial to get a wider range of participants, or just a larger number of them than this paper had. It may give a better idea as to who is willing to share information and tell why.

Running the Asylum - first 7 chapters

Book: The Inmates are Running the Asylum

Author: Alan Cooper

Comments: Jill, Aaron

Summary:

In the first 7 chapters Alan Cooper is trying to show that interaction design makes people frustrated, and that it needs to be changed in order to make it more "user friendly". Many users feel stupid when they can't control their computers, and the way Alan Cooper says it "programmers don't think about the user" and think that most things are user error. I wasn't really sure what to think of this book, but there were a few things that he mentioned that did make sense.

• Things you combine with a computer end up just being a computer. It doesn't necessarily make anything easier to use, and in fact it usually makes the item harder to use than it was originally intended.
• Cognitive Friction: "the resistance encountered by human intellect when it engages with a complex system of rules that change as the problem changes" (19).
• Wasting Money: money is wasted when programs have a deadline and the programmers don't have enough time to implement and add functions that are necessary for a good product. Being late with a product isn't as bad as developing something that doesn't work that well. Money is wasted this way though, and all the management wants is more features and a program that they think will sell.
• The Dancing Bear: features that seem to work well, and people are unaware that the feature could actually be much better than it is.
• Customer Disloyalty: because programs are being produced that aren't to their full potential, companies are losing clients and they aren't buying their products anymore.

Discussion:

I think Cooper is trying to show that programmers should not also be in charge of the design, and if design would go first then the programming would be better. This would save money, and produce better features that weren't just "dancing bears". I felt as though he was picking on programmers, especially in chapter 7 when he compares programmers to jocks. But that scenario can be rather true when really thought about. I believe that if the design was implemented first, then programming would be better and the programmers would actually be able to shine more by making the functions actually work. Since the world is so demanding time limits and constraints are what makes some interaction design become much more frustrating to the user than it was intended. Some things should really not be released until they are fully functional and truly "user friendly".

Design of Haptic Interfaces for Therapy

Written by: Cati Vaucelle, Leonardo Bonanni and Hiroshi Ishii

Comments: Zach

Summary:

This paper deals with devices that may be able to be used for therapeutic purposes. Therapeutic holding has been something that has worked for a few disorders because it feels like hugging. One case would be an autism child wearing a heavy blanket to stop them from having panic attacks. They came up with four different prototypes.

1. "Touch Me": a flexible motor array in a soft enclosure that would be used to touch certain parts of the body for therapeutic purposes.
2. "Squeeze Me": would be use to promote human contact with patients who would get to overwhelmed if having to deal with a real person. The hugging-like sensation could keep them from having panic attacks, or other issues that they deal with with their disorders. It could also be worn outside of a clinic area without people taking notice to it because the pressure bladders go inward towards the body and should not be noticeable under a shirt.
3. "Hurt Me": this would be used for those patients that usually mutilate themselves when dealing with a disorder that they have. This would give them the pain sensation that they need without actually hurting them self. This would also give a closer relationship between therapist and patient so the therapist realizes the pain their patients feel and need to be ok.
4. "Cool Me Down": has a temperature all the way around the wrap that will help patients self-administer a cooling sensation for sensory grounding.

These prototypes were presented to a few therapists and they gave their approval that these devices would be good to test on patients as long as there were safety measures being met. The "cool me down" may not be able to be very portable due to the motor needed to run it, but the others would be able to. They also thought "Hurt Me" would need a new name in order to be used. Now they just need to actually test these prototypes on patients in order to see if they actually work as intended.

Discussion:

I think this is a great idea, especially the "Hurt Me" so people will not mutilate their bodies in order to feel the pain they need with their disorder. If these therapeutic prototypes will work for certain people with disorders to keep them from having panic attacks or violent tendencies then I think this is something we should begin working on. Future work would obviously be actually testing these things on patients and recording whether or not it actually worked for them. It's something to really think about and hopefully be able to produce so people with disorders can qualm their needs without harming themselves or others.

It's a GrayArea...

Paper: It's Not That Important: Demoting Personal Information of Low Subjective Importance using GrayArea

Written by: Ofer Bergman, Simon Tucker, Ruth Beyth-Marom, Edward Cutrell, and Steve Whittaker

Comments: Randy

Summary:

Everyone who owns a computer has to deal with organizing their data so that they can find it whenever they need it. For some users it is hard to decide whether or not to delete an item, or keep it because it may be used again. GrayArea is something that will allow people to demote files that they can't decide if they will need again or not into a gray area that is part of the explorer. When searching for something on their computer users may feel guilty for being "unorganized" and it will distract them from opening the file they had originally went to look for. This group of people created three different designs that they asked for feedback before creating a prototype.

1. Chronological Sorting: listing documents from recently viewed instead of the Microsoft way of viewing by recently changed. This way documents that have not been open in a very long time would be on the bottom, and the ones viewed/used a lot would be on the very top.
2. FadedItems: would give the user an ability to fade out the files that they thought were not very important. The more important ones would remain completely visible. (Fade and Unfade options)
3. GrayArea: allow users to drag files and folders into a designated gray area so that they would not have to make the decision to delete them for good.

There were a few things that they had to consider when creating a prototype.

• Direct or Auto manipulation: should the user be able to determine what files and folders to manipulate (fadeditems and grayarea), or should the computer do that for them (chronological ordering)
• Designated or Undesignated location: should they remain in the same folder and place, or should they be moved somewhere else where they will be less of a distraction.
• Continuous or Categorical demoting

There were 79 participants that were polled on what they thought would be a design that they would use. They answered questions and responded with an answer from 1-5 on whether they liked the idea or not. The results are listed below.

They considered these results and the reasons behind why people chose what they did. For Chronological Sorting most people didn't even use the chronological sorting already implemented in the operating system, so why edit it if people haven't used it and probably won't. For FadedItems the folders would soon become crowded with hidden files which would not take away the distracting unwanted folders when searching for something. So with this logic they decided to make the GrayArea prototype and allow users to test this out. They made a working Java prototype that looked and felt similar to the Explorer we see in Windows. They didn't want to fuss with actually extending explorer at the time because of all the loops they would have to jump through. With the working prototype users were told how to work GrayArea, and were given two tests. One test would be to go through their folders and either delete or keep files without the use of GrayArea. Then they were asked to clear their folders with GrayArea where you could demote some to the gray area at the bottom of the explorer or delete/keep files the normal way. Once the tests were over, they gathered their data to make a chart for comparison.

This chart shows that Gray Area has helped with the ease of keeping or deleting folders for people because they do not feel pressured into deleting a file they might never need. They could just demote the file into the GrayArea and have the ability to get them if ever needed.

Discussion:

I think this would be something I would use because there are always certain folders or files that I have that I'm unsure if I should delete or not. There are also a lot of classwork files that I want to keep just in case I ever need them, but they clutter my hard drives so it is sometimes hard to find whatever it is I was originally looking for. They hope to create this for all operating system platforms, but I just hope that Apple and Windows don't get cranky and refuse to let them do it. I think it would be a great addition to our operating systems and could help improve the way we organize our folders. Future work would consist of writing this prototype in different platform languages so that it would fit with each type of operating system. It will also probably take a lot of persuading to get those companies to let them do what they want. I for one would look forward to something like this, and hope they get a chance to do it.