I was a volunteer in ITS. I've listened to lots of good talks and attended some great workshop sessions.

I think the idea of getting people who work for different aspects of ITS together is great. With a educational data mining background, I learned better about the architecture of the system. I especially like the idea of collaborative system.

What I learned from ITS2010 about how to frame, plan, perform, and communicate my own ITS research

I am a PhD student in Computer Science. I work in Human Computer Interaction.

My work is about distance tutoring system where we use a human-driven avatar for embodied interaction. Our focus is on multimodal communication and believable behavior generation.

This was the first time I attended ITS conference and I had a great experience. Not only did I have a chance to present my works as a poster, but I also met other researchers in ITS community, a community I wasn’t familiar with. That was my main goal when I attended to this conference, to understand what they do. Even though I am working on a distance tutoring system where we have a tutor, our approach is different. And that was the missing part in my expertise.

I really enjoyed interactive events during Opening reception. It makes it much more easier to understand different systems developed by researchers from different areas. I think I not only learned from these researchers but I also contributed to their work by giving my feedback. I have a different background, and realized that some of the people find it valuable.

There was an interesting presentation titled “Examining the Role of Gestures in Expert Tutoring” by Williams. I had a chance to discuss my work with her. It was really nice to meet someone who is working on a very similar project with different perspective. Our final goals are similar, but we started from different parts and we focused on different parts. It was a nice discussion and we exchanged ideas.

Another thing I liked about ITS is the number of invited speakers. Having five speakers is not that usual for a conference of this size. I found them mind opening. Each brought different perspective. I was inspired by Beverly Park Woolf in many ways.

I just wish that more people visited the room where I presented my poster. It was the last room with around six posters, and a lot of people just went to the first one across the room where they picked their lunch. Only a few people came to ours. I am glad that I informed some of the people I met to come and see it. They came and we had a nice discussion.

Overall, I learned a lot. I will apply some of the things I learned in ITS to my future research. I am definitely planning to attend ITS 2012.

This conference was my first exposure to a lot of the research in the ITS area. I attended as many talks as possible, and several stood out to me in particular.

In the talk entitled Characterizing the Effectiveness of Tutorial Dialogue with Hidden Markov Models, Dr. Boyer et al. wanted to explore effective tutorial dialogue strategies in order to build a data-driven tutorial dialogue system. They used an HMM to model these dialogue strategies and help predict what a human tutor would do in a problem state, since human tutors are consistently more effective than computerized tutors. To do this, they viewed each tutoring session as a series of interactions, and created a set of dialogue act tags that captured the purpose of each dialogue move within the tutoring session (such as assessing questions, statements, feedback, etc.). Because an HMM has an initial probability distribution, they had the starting state, and because the HMM also has a transition probability distribution, they could determine the likelihood that it would move to another state. Using these HMMs, they were able to evaluate the tutoring modes without supervision, and found that student learning is correlated with automatically extracted tutoring modes, when the tutoring structure was learned from tutoring sessions. I thought it was very interesting that these HMMs could provide a high-level structure of the progression through a tutoring session, and that they were also correlated with student learning.

In Exploiting Predictable Response Training to Improve Automatic Recognition of Children‟s Spoken Responses, Chen et al. built on the existing Project LISTEN to create a reading tutor that teaches the reading comprehension strategy of self-questioning. Since the target population is children in grades 1-3, they cannot type yet. The tutor reads the question out loud, and the children speak their responses. One thing that was really cool to me was that it looked at the children‟s responses from the standpoint of “if you know what the speaker is going to say, it makes it easier to listen.” The system narrowed down what the children might ask by providing a question stem and a character to ask about in the story. It used finite state automata to build the question: stem + character + completer. Since children often repeat themselves when trying to voice a question, the FA can handle this by just sending it back to the beginning state if the child is repeating. I thought this was a really elegant solution – an „aha‟ moment – and I was really excited to see it employed in the context of reading. Also, even if the system could not figure out the child‟s question, it could still use its vocabulary from 673 children‟s stories in order to fill in the child‟s question. If none of the words match, it is most likely off-task speech. It will be interesting to see where this research goes in the future – tutors that can listen to children and respond accordingly would be very valuable.

In The Impact of Gaming(?) on Learning, Gong et al. examined how much gaming behavior impacts learning in ITS environments (there are effectively zero learning gains when gaming), and looked at novel predictors of gaming. Students game on skills in which they are relatively weaker compared to other areas, which means that the identity of individual students must be preserved in order to detect gaming behavior. The basic gaming detector identified rapid guessing (submission of answers less than two seconds apart), rapid response (the performance of any action after a hint or starting a problem before a reasonable amount of time had passed to allow for reading the question), and repeatedly exhausting all hints for questions. They then used the performance data of students to make boolean decisions about whether a student was gaming for each given problem. To do this, they employed an HMM, where student performance was the observed variable and student knowledge was the latent attribute. The HMM used the initial knowledge to move through the states of learning. They added a gaming model that played into the states as well, so that the states would be traversed differently if the students were gaming, and they may or may not learn. They combined this with a contextual model for smoothing based on past data, and created a skill model that could calculate the learning and gaming rates for students across 106 skills. They found that low knowledge for gaming is relative, and there is apparently no absolute threshold for gaming – students just tend to game on their weaker skills. Some students tend to game a lot, while other times difficult problem sets or skills that have not yet been taught in class can be the cause of gaming. Overall, gaming is influenced by the student, and there are orders of magnitude of difference in the learning gains for students who game versus those who do not.

Hello, I am a second year PhD student on Computer Science and I am working on Affective Tutoring Systems development…. My work is about affective computing, software engineering and software architecture.

This was the first time I attended ITS conference and I think the experience worth it. I participate at one hand, as student volunteer and at the other hand as tutorial speaker. Both activities let me meet and interact with different members of the ITS community.

AS STUDENT VOLUNTER, I learned more about the efforts and initiatives of the ITS community mainly about the implications of affective elements on the learning process, a topic in which I am focusing my effort in order to make synergy and create new opportunities for new projects. In this area I have the opportunity to talk and share with speakers, poster presenters and other student volunteers.

In particular I had the chance to meet a Canadian student (he was assisting me as student volunteer during my tutorial) and we shared common interests in project topics and entrepreneur initiatives. We had very rich conversations where we found synergy opportunities and shared research development. For sure we will be working together soon!

Beside that, it was positive surprising for me to find people from Latin America (LA) who are actively working in the ITS community. That is because besides being a PhD student at Arizona State University, I am part of the Latin American faculty. During ITS we was able to create a small--‐but--‐full--‐of--‐ideas--‐group with people from Mexico, Brazil, Costa Rica and Spain. Our goal as a group will be to create initiatives that could positively impact the different regions we represent in LA.

AS TUTORIAL SPEAKER, I shared my experience with the tutorial named “How to Apply Software Architecture and Patterns to Tutoring System Development?”. This talk was motivated because during my research work, I had found that there is some level of misunderstanding and misconception about software engineering and software architecture processes inside the ITS community.

It was an interesting talk about different concerns on the community such as cost, time of development, flexibility, modularity, robustness, data log, and creation of distributed applications, and share ideas about the best way to apply software architecture principles and mainly try to explain to the attendees how having a common language and a common set of tools helps and facilitates activities of both, the authors and the developers. Besides sharing knowledge, ideas and experiences, I was interested on learning more about how ITS community conceives the whole software engineering process, and how to better clarify misconceptions. I believe this will help me to guide my research about them. I submitted a paper to ITS 2010 conference, it was rejected, however now I know that the software engineering vocabulary is pretty different from the one used among ITS community; terms such as module, component, client have a different meaning inside ITS community compared with software engineering community. Definitively the interchange of ideas generated during the tutorial will positively impact my writing guidelines in order to be able to share my work among the ITS community.

Summarizing, this was a really good opportunity to get involved with the ITS community, create collaboration opportunities, share ideas and rethink my writing guidelines.

Being this the first time I attend ITS conference it was a very good experience. All my background is on Computer Science (undergrad and master) now that I am pursuing my PhD on Computer Science and I am working on projects related with ITS community, particularly on tutoring systems and affective computing.

One particular thing that was grateful surprising is learn about the different efforts and initiatives that are growing on different places on Latin America, this conference help us (Latin American community) to get closer and share projects that positively impact our regions in joined projects with the rest of the ITS community efforts.

As part of my PhD work, I am contributing with the development of an affective ITS. For me it is important to learn more about the point of view of the ITS community (being my background on software engineering) regarding software development.

I had the chance to participate as a tutorial co-speaker, and share with the community a session about How to Apply Software Architecture and Patterns to Tutoring System Development?. My main interest on giving this tutorial was learn more about the concerns of the ITS community and get a better sense of how I could positively contribute and impact in this area.

I consider that the session was worth it, due the fact we can talk about different concerns on the community, and how the use of software architecture and patterns could help to obtain better applications that better suits the ITS community. I also find interesting that the session allow us all to talk and clarify about some misconceptions in both sides (ITS and Software communities) and then help us to find and define a common ground to build on the top of it in the future.

I am very thankful to the ITS committee for awarding me a scholarship to attend ITS 2010. The conference was very educative and useful for me.  The minute madness presentation forced me to learn how to present my research in an engaging manner in one minute and on one slide.

I also learned that sharing my research ideas early through a poster is important because I can get feedback and literature review for free. People tell you research they have seen or heard about that is similar to what you are doing.

The question generation workshop made me realize that I wasn’t working in a saturated field and that in fact, there is a lot more research work to be done in question generation.

Most important of all, I had spent a couple of weeks looking for a free semantic role labeler to no avail. I talked to a couple of people at the conference and suddenly, I got inundated with a lot of recommendations.  These recommendations were my biggest score at the conference. But, I didn’t just get recommendations, the recommenders also offered to help me if I had problems installing, using or retraining the semantic parsers.

In addition, I was greatly inspired by the invited talks, especially the talk by Dr. Chee-Kit Looi. I like how they are integrating the product of education research into classrooms on such a somewhat large scale.

This was my first ITS conference and I’m glad I was able to attend.  Now, I feel like I belong to the ITS community. I met a lot of the renowned researchers in intelligent tutoring systems such as Kurt Van Lehn, l met people whose papers I had read and cited and I met up and coming researchers just like myself.

Seeing a lot of women and other minorities at the conference also increased my feeling of belonging to the ITS community.

Once again, I am thankful to the ITS committee for giving me the opportunity to experience the ITS conference and to meet, network and share research ideas with other ITS researchers.

What a conference…  This was my third ITS conference and it was an incredibly formative and informative experience.  I watched several themes unfold that help me look at the ITS community and the challenges we face in a different way. In addition, the organizers and other senior community members provided a glimpse of the inner-workings of the research system that is incredibly helpful in understanding how to work more effectively as a community member myself. To make the experience round-trip, it was also very rewarding to see others who have been influenced by my earlier work and are now learning from the lessons that I have put so much effort into sharing.

The total of the content I have attempted to absorb during the week would be impossible to share in this short note, but there was one surprising message upon which I find myself reflecting. The keynote speaker Dr. Chee-Kit Looi stated the emergent theme eloquently: We need to bridge the gap between research and the classroom.  This is an old and obvious idea in some sense, but many people brought it into a different perspective for me at this conference. Researchers are starting to look at this question as a research area, rather than a practical roadblock.  A perfect example, and a very interesting piece of work, is Dror Ben-Naim's efforts to return "pedagogical ownership" to the teachers. Taking this type of research-based approach to bringing teachers into our tutor-creation process might be the way to find real solutions to this problem.

In addition to the information about tutoring systems themselves, this conference was also very informative about the community in which they are created.  This started with Dr. Kay's welcome talk where she gave us a clear idea of how reviews are put to use and what the process of selecting papers for publication is like. The next major enlightening moment about process was the panel discussion when Marco Marsella and John Cherniavsky discussed details about the goals and achievements of the funding agencies.  Finally, Dr. Mostow, through his presence and humor during the conference, let us all know a little about what it must be like to organize a gathering of this scale.  It is rare to get such a glimpse into the way our community works from the perspective of different stakeholders: deciding funding, accepting papers, even organizing the forum in which ITS builders can share information and grow.  The transparency and level of detail that was shared helped me learn much about how a research community operates.

Finally, when attending this year, I noticed that for all the ITS community has done for me; I have given back to the group as well.  I first contributed to ITS in 2006, and a fellow researcher, Ido Roll, approached me during this conference to tell me that he had employed the tactics presented in my first work to build the system he was presenting this year. Hameedullah Kazi presented a system that had similarities to our work, and when discussing it with him he said that he had used my prior publications as reference.  Finally, Niels Pinkwart presented a poster that showed survey results from experts on developing argumentation systems.  I participated in that survey by invitation. So, taking all of this into consideration in just one short week, I can see how my contributions are starting to be of use to a larger group of researchers.

This year's ITS conference helped me grow as a research scientist in many facets.  I am grateful to be able to attend such a gathering.  It has given me inspiration, capability, and confidence to approach and contribute to the exceptional body of work that ITS community provides.

From my own perspective, this experience is very good for strengthening ties between individuals that will make the ITS community of tomorrow. Networking is maybe the most important aspect of attending ITS2010 conferences as a young researcher. Furthermore, during this week, we had several off-conference events where we learnt to better know each others. I have had the opportunity to read about the work of many of other ITS attendees. But knowing the individuals behind the papers sometimes provides good insight to fully understand the motives and the future of their respective project. This definitely has an important impact on my own thinking, which I can also improve by confronting it to perspectives of others in an caring and friendly setting that email-based discussions would not reproduce.

One of my main research interests is about considering cultural specifics of individuals in ITS environments. People I met there were from different origins, refer to different systems of values, perceptions of their environment, and sometimes have a very different opinion about what a “good education” and a “good learning session” are. Still they share an interest for Intelligent Tutoring Systems, and in general are above average open-minded people (I think). For me as a researcher interested in culture, it was thus a great opportunity to share my thought, to interact with people, and then to see how this had different impacts (and I had the opportunity to discuss the reasons of such variations, and to look for potential cultural explanations). This vicarious learning experience was also a very important aspect of my ITS2010 experience. 

I have not experienced any “eureka” moments during the ITS conference. However, since I see “thinking” as a maturing process, all the great discussions I had the opportunity to have by attending the ITS conference will definitely have a profound impact on my future research production, as well as on collaborations I will strive to create/develop with people I met there.