user2code2vec

Abstract

In this work, we propose a new methodology to profile individual students of computer science based on their programming design using a technique called embeddings. We investigate different approaches to analyze user source code submissions in the Python language. We compare the performances of different source code vectorization techniques to predict the correctness of a code submission. In addition, we propose a new mechanism to represent students based on their code submissions for a given set of laboratory tasks on a particular course. This way, we can make deeper recommendations for programming solutions and pathways to support student learning and progression in computer programming modules effectively at a Higher Education Institution. Recent work using Deep Learning tends to work better when more and more data is provided. However, in Learning Analytics, the number of students in a course is an unavoidable limit. Thus we cannot simply generate more data as is done in other domains such as FinTech or Social Network Analysis. Our findings indicate there is a need to learn and develop better mechanisms to extract and learn effective data features from students so as to analyze the students' progression and performance effectively.

Evaluating Distance Measures for Program Repair

“…We refer to this dataset as "The Dublin data" since it contains data collected at Dublin City University. The original dataset released by Azcona et al [4] contains more than half a million programming submissions (591,707) by 666 students from five Python programming courses over three academic years. The assignments vary in difficulty levels ranging from basic printing to more complex sorting algorithms.…”

Improving learning experiences in software engineering capstone courses using artificial intelligence virtual assistants

“…In the application of tools based on Artificial Intelligence to support the teaching of software engineering, Azcona [4] implemented a Chatbot using AI called CoderBot. This bot encourages students to get involved and be more proactive in their learning.…”

Clustering Introductory Computer Science Exercises Using Topic Modeling Methods

“…Similarly, code clustering, identifying similar features/concepts in a repository, predicting bugs, analyzing/predicting source code evolution, tracing links between modules, and detecting clones are widely applied in the same field [31]. For the EDM, Azcona et al [32] created a Python code submission tokenizer to setup features to generate code embeddings. Unlike the software engineering context, code snippets in CS1 are small in size and lack annotations.…”