Papers by Jeffri Murrugarra-Llerena
Improving Embeddings Representations for Comparing Higher Education Curricula: A Use Case in Computing (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to represent study programs using bag-of-words and clustering algorithms are prone to biases due to personal beliefs and perspectives. |
| Approach: | They propose to use pre-trained word embeddings to fine-tune a study program classification task to obtain more accurate curriculum representations than strong baselines. |
| Outcome: | The proposed method is compared to existing methods on a study program classification task and on comparing computing curricula from USA and Latin America. |