Week 27: 3/29/17 - 4/5/2017

Work Accomplished: 
This week, I finished the list of Keyword AOIs from the 9 tasks in our Stack Overflow study. I sent them to Jenna to check, but I found it easier to do a manual analysis than waiting for the data to export. I also presented 'Improving Stack Overflow Tag Prediction', the first half of our study, at Youngstown State University's QUEST, a forum for undergraduate and graduate research. Jenna and I presented a poster. We are also using Databricks, a virtual analytics platform, to run Apache Spark. I uploaded the train data from the Kaggle competition as a cluster and I'm going to run a bigger file (also part of the Kaggle Competition) to see how the two compare against one another. We are then going to take our participant fixation data and run it in the online forum. 

Weekly Goal: Get Alex scheduled to participate in the study. 

Future Goal: Start writing the abstract for the second half of our study to be submitted. Finish the data analysis. 

Week 26: 3/22/2017 - 3/29/2017

Work Accomplished

This past week I finalized my poster for QUEST titled, Towards Mining Eye-tracking Datasets for Expertise Prediction, and sent it to printing. I also finished the R script to complete step 3 of the data pre-processing steps that merges rows in each task's dataframe based on the fixation index. The hardest part about writing this R script was selecting the correct merging pattern for each data field across each dataset. I also completed a fellowship application to our Phi Kappa Phi chapter at YSU.

Goal

Weekly goal(s) - In the next week, I will be double checking keywords that Alyssa makes for the non-eyetracking keyword selection. I will also be working to gather statistics on the new datasets I created, such as the number of views of each AOI per task per participant and across participants. This should help us narrow down keywords.
Long-term goal(s) - Predict keywords by modifying the process of a propose method for keyword prediction in a Kaggle competition by incorporating eye-tracking. We will also predict keywords without eye-tracking and compare the two keyword sets generated. These keywords inform our tags, so determining them will tell us which pieces of code and/or text in a StackOverflow document are pertinent to tag selection.

Outcome(s)

1. QUEST poster sent to printing and final draft completed
2. R script to complete step 3 of the data pre-processing steps written (see above explanation of step 3)
3. Phi Kappa Phi Fellowship application completed

Week 26: 3/22/2017 - 3/29/2017

Work Accomplished: 
This week, I started working on creating keywords AOIs using the data we've collected for our study. I manually examined similar keywords for all 16 participants and determined each one for the 9 tasks. I also uploaded my pre/post survey data to Dropbox under Survey Results. The next step is to take the Train and Test data from a Kaggle competition on Facebook and run it through Apache Spark to test for a keyword analysis. Then, I will take our data and run it through to see if we get similar results. After I export the data and create an excel file with images corresponding to our AOIs, Jenna will run that data through her R script to separate each value. From there, we will analyze the data further and come up with our tag prediction system.


Weekly Goal: Get up to 20 participants for the study. I was able to get 1 more person scheduled, but they were unable to do the study last week.


Future Goal: Take the exported data and compare it against the train and test data from predicting keywords on Kaggle.

Week 25: 3/15/2017 - 3/22/2017

Accomplishments: 
This week, I wrote up the abstract for my poster at QUEST, Improving Stack Overflow Tag Prediction Using Eye Tracking and Jenna submitted it for me. I also gathered the new participant data and uploaded the pre/post responses to our Dropbox. Next week, I will be in charge of creating keyword AOIs in Tobii Studio and hopefully gathering the last of our remaining participants. 

Weekly Goal:  Gather more participants for our study.

Future Goal: Run data through machine learning algorithm and present my poster at QUEST in April. 

Week 25: 3/14/2017 - 3/22/2017

Work Accomplished

This past week I created a draft of my poster for QUEST titled, Towards Mining Eye-tracking Datasets for Expertise Prediction. I also finished the R script to complete steps 1 & 2 of the data pre-processing steps listed below. The hardest part about writing this R script was splitting up the data into files with all participant data per task, because the data is currently per participant where each participant file has the eye-tracking data for all tasks.

Goal

Weekly goal(s) - In the next week, I will be working on step 3 of the data pre-processing steps below, which will be the most difficult of all of the steps, because aggregations are tricky. I will also be presenting the draft of my poster for QUEST as practice for the actual presentation.
Long-term goal(s) - In the next few weeks, I will be working to accomplish the pre-processing steps listed below. (2 of which are already complete)

1. Create one excel file with the TobiiStudio Field name shown on the left for each participant with all tasks
2. Create a file for each task with all participant data with relevant experiment field name data columns
3. Create a file for each task with fixation and duration data merged. All other columns remain as in step 2.

Outcome(s)

1. QUEST poster draft completed
2. R script to complete steps 1 & 2 of the data pre-processing steps written

Week 24: 3/8/2017 - 3/14/2017

Work Accomplished

This past week I worked on writing and submitting my abstract for a poster to QUEST at YSU. My poster will be about mining eye-tracking data using sequential analysis techniques; a presentation of the paper we submitted to MSR a few weeks ago. I also spent time touching base with Dr. Sharif (outside of our regular meeting time) to discuss next steps for our StackOverflow data analysis and I got a clearer picture of the data pre-processing steps we need to complete in the coming weeks:

1. Create one excel file with the TobiiStudio Field name shown on the left for each participant with all tasks
2. Create a file for each task with all participant data with relevant experiment field name data columns
3. Create a file for each task with fixation and duration data merged. All other columns remain as in step 2.

I will be working with Alyssa in the next few weeks to accomplish these steps.

I also found out yesterday that I will be receiving the 2017 NSF Graduate Research Fellowship. So, that was exciting news!!

https://www.fastlane.nsf.gov/grfp/AwardeeList.do?method=loadAwardeeList

Goal

Weekly goal(s) - In the next week, I will be working on step 1 of the pre-processing steps listed above and on my poster for QUEST.
Long-term goal(s) - In the next few weeks, I will be working with Alyssa to accomplish the pre-processing steps listed above.

Outcome(s)

1. Clear next steps for data pre-processing for the StackOverflow project
2. QUEST poster abstract submission

Week 24: 3/8/2017 - 3/16/2017

Work Accomplished: 

This week, I had to write an R script with the help of Jenna. I am still working on the R script that will be able to process our excel files once I export the data from Tobii Studios. I've collected user pre and post questionnaire data for the users who participate in our Stack Overflow Eye Tracking study. I have those files stored on my computer and I have to upload them to our Dropbox. I also wrote an abstract for YSU's QUEST in April. I attended a digital media conference in New York City over spring break and it gave me perspective from software developers working in newsrooms. The outbreak of digital media will continue to grow and it was interesting to see the software that some of the outlets use to track data. Storyful, a news outlet that gathers data from social platforms around the world, uses a heatmap to store and track news articles/video from around the world. It's refreshing to know that even traditional media outlets are starting to realize the importance of online content and data technology. 

Work Accomplished: Collected more data, submitted QUEST abstract for review. 

Future Work: Present poster at QUEST, finish R script to export user data and submit pre/post questionnaire data to Dropbox.