Use expert knowledge instead of data

generating hints for hour of code exercises

Milo Buwalda, J.T. Jeuring, Nico Naus

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

Abstract

Within the field of on-line tutoring systems for learning programming, such as Code.org's Hour of code, there is a trend to use previous student data to give hints. This paper shows that it is better to use expert knowledge to provide hints in environments such as Code.org's Hour of code. We present a heuristic-based approach to generating next-step hints. We use pattern matching algorithms to identify heuristics and apply each identified heuristic to an input program. We generate a next-step hint by selecting the highest scoring heuristic using a scoring function. By comparing our results with results of a previous experiment on Hour of code we show that a heuristics-based approach to providing hints gives results that are impossible to further improve. These basic heuristics are sufficient to efficiently mimic experts' next-step hints.
Original languageEnglish
Title of host publication L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale
Place of PublicationNew York, NY
PublisherAssociation for Computing Machinery (ACM)
ISBN (Print)9781450358866
DOIs
Publication statusPublished - 26 Jun 2018
EventFifth Annual ACM Conference on Learning at Scale - London, United Kingdom
Duration: 26 Jun 201828 Jun 2018
https://learningatscale.acm.org/las2018/

Conference

ConferenceFifth Annual ACM Conference on Learning at Scale
Abbreviated titleL@S 18
CountryUnited Kingdom
CityLondon
Period26/06/1828/06/18
Internet address

Fingerprint

Online systems
Pattern matching
Computer programming
Students
Experiments

Cite this

Buwalda, M., Jeuring, J. T., & Naus, N. (2018). Use expert knowledge instead of data: generating hints for hour of code exercises. In L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale [32] New York, NY: Association for Computing Machinery (ACM). https://doi.org/10.1145/3231644.3231690
Buwalda, Milo ; Jeuring, J.T. ; Naus, Nico. / Use expert knowledge instead of data : generating hints for hour of code exercises. L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale. New York, NY : Association for Computing Machinery (ACM), 2018.
@inproceedings{8f14789c5b8647f9a3a3d8b2800737fa,
title = "Use expert knowledge instead of data: generating hints for hour of code exercises",
abstract = "Within the field of on-line tutoring systems for learning programming, such as Code.org's Hour of code, there is a trend to use previous student data to give hints. This paper shows that it is better to use expert knowledge to provide hints in environments such as Code.org's Hour of code. We present a heuristic-based approach to generating next-step hints. We use pattern matching algorithms to identify heuristics and apply each identified heuristic to an input program. We generate a next-step hint by selecting the highest scoring heuristic using a scoring function. By comparing our results with results of a previous experiment on Hour of code we show that a heuristics-based approach to providing hints gives results that are impossible to further improve. These basic heuristics are sufficient to efficiently mimic experts' next-step hints.",
author = "Milo Buwalda and J.T. Jeuring and Nico Naus",
year = "2018",
month = "6",
day = "26",
doi = "10.1145/3231644.3231690",
language = "English",
isbn = "9781450358866",
booktitle = "L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale",
publisher = "Association for Computing Machinery (ACM)",
address = "United States",

}

Buwalda, M, Jeuring, JT & Naus, N 2018, Use expert knowledge instead of data: generating hints for hour of code exercises. in L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale., 32, Association for Computing Machinery (ACM), New York, NY, Fifth Annual ACM Conference on Learning at Scale, London, United Kingdom, 26/06/18. https://doi.org/10.1145/3231644.3231690

Use expert knowledge instead of data : generating hints for hour of code exercises. / Buwalda, Milo; Jeuring, J.T.; Naus, Nico.

L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale. New York, NY : Association for Computing Machinery (ACM), 2018. 32.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

TY - GEN

T1 - Use expert knowledge instead of data

T2 - generating hints for hour of code exercises

AU - Buwalda, Milo

AU - Jeuring, J.T.

AU - Naus, Nico

PY - 2018/6/26

Y1 - 2018/6/26

N2 - Within the field of on-line tutoring systems for learning programming, such as Code.org's Hour of code, there is a trend to use previous student data to give hints. This paper shows that it is better to use expert knowledge to provide hints in environments such as Code.org's Hour of code. We present a heuristic-based approach to generating next-step hints. We use pattern matching algorithms to identify heuristics and apply each identified heuristic to an input program. We generate a next-step hint by selecting the highest scoring heuristic using a scoring function. By comparing our results with results of a previous experiment on Hour of code we show that a heuristics-based approach to providing hints gives results that are impossible to further improve. These basic heuristics are sufficient to efficiently mimic experts' next-step hints.

AB - Within the field of on-line tutoring systems for learning programming, such as Code.org's Hour of code, there is a trend to use previous student data to give hints. This paper shows that it is better to use expert knowledge to provide hints in environments such as Code.org's Hour of code. We present a heuristic-based approach to generating next-step hints. We use pattern matching algorithms to identify heuristics and apply each identified heuristic to an input program. We generate a next-step hint by selecting the highest scoring heuristic using a scoring function. By comparing our results with results of a previous experiment on Hour of code we show that a heuristics-based approach to providing hints gives results that are impossible to further improve. These basic heuristics are sufficient to efficiently mimic experts' next-step hints.

U2 - 10.1145/3231644.3231690

DO - 10.1145/3231644.3231690

M3 - Conference article in proceeding

SN - 9781450358866

BT - L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale

PB - Association for Computing Machinery (ACM)

CY - New York, NY

ER -

Buwalda M, Jeuring JT, Naus N. Use expert knowledge instead of data: generating hints for hour of code exercises. In L@S '18 Proceedings of the Fifth Annual ACM Conference on Learning at Scale. New York, NY: Association for Computing Machinery (ACM). 2018. 32 https://doi.org/10.1145/3231644.3231690