ITM-Rec: An Open Data Set for Educational Recommender Systems

[Data Description]

ITM-Rec data sets including student and group preferences on the topics of final projects in data base and data science classes, and it was collected from student questionnaires at the ITM department, Illinois Institute of Technology, USA. It can be utilized for the purpose of developping and exmining different educational recommender systems (RS), e.g., context-aware RS, multi-criteria RS, group RS, and RS with integrated information, such as multi-criteria based context-aware RS, context-aware based group RS, and so forth.

For more information, you can refer to the two publications below:

 - Yong Zheng. "ITM-Rec: An Open Data Set for Educational Recommender Systems". Companion Proceedings of the 13th International Conference on Learning Analytics & Knowledge (LAK), Arlington, TX, USA, March 13-17, 2023

 - Yong Zheng. "Personality-Aware Decision Making In Educational Learning", Proceedings of the 23rd ACM Conference on Intelligent User Interfaces (ACM IUI), Tokyo, Japan, March 7-11, 2018

Notes: Students' personality traits were not disclosed in this data set, due to privacy concerns.

[Citations]

If you used this data for the purpose of research, please cite the following publications:

 - Yong Zheng. "ITM-Rec: An Open Data Set for Educational Recommender Systems". Companion Proceedings of the 13th International Conference on Learning Analytics & Knowledge (LAK), Arlington, TX, USA, March 13-17, 2023

[File Descriptions]

 - users.csv

   columns: UserID, Gender, Age, Married

   description: Meta data about students

 - items.csv

   columns: Item, Title, URL, Descriptions

   description: Meta data about the topics of projects

 - ratings.csv

   columns: UserID, Item, Rating, App, Data, Ease, Class, Semester, Lockdown

   description: Students' individual ratings on items, including the overall rating (i.e., the column 'Rating'), and multi-criteria ratings (i.e., ratings in 'App', 'Data', 'Ease'), as well as three contextual variables, e.g., class, semester, lockdown

 - group.csv

   columns: GroupID, UserID

   description: the compositions of groups

 - group_size.csv

   columns: GroupID, Size

   description: the number of students in each group

 - group_ratings.csv

   columns: GroupID, Item, Rating, App, Data, Ease, Class, Semester, Lockdown

   description: Ratings on items given by groups, rather than individuals

[Data Statistics]

Ratings by individual students:

# of users: 476

# of items: 70

# of ratings by individual users: 5230

distribution of rating amounts by users: {mean: 11.52, min: 3, max: 51, std: 7.06}

distribution of rating stars (Rating): {1: 724, 2: 1050, 3: 555, 4: 1348, 5: 1553}

distribution of rating stars (App): {1: 514, 2: 862, 3: 1062, 4: 1492, 5: 1300}

distribution of rating stars (Data): {1: 421, 2: 872, 3: 1284, 4: 1549, 5: 1102}

distribution of rating stars (Ease): {1: 474, 2: 900, 3: 1713, 4: 1511, 5: 632}

Ratings by student groups:

# of groups: 143

# of items: 70

# of ratings by groups: 1117

group size: {2: 88, 3: 42, 4: 9, 5: 4}

distribution of rating amounts by groups: {mean: 7.81, min: 5, max: 22, std: 2.47}

distribution of rating stars (Rating): {1: 115, 2: 306, 3: 172, 4: 2244, 5: 280}

distribution of rating stars (App): {1: 87, 2: 229, 3: 281, 4: 292, 5: 228}

distribution of rating stars (Data): {1: 63, 2: 227, 3: 309, 4: 333, 5: 185}

distribution of rating stars (Ease): {1: 78, 2: 205, 3: 344, 4: 355, 5: 135}