XES3G5M

XES3G5M is a knowledge tracing benchmark dataset collected from TAL Education Xueersi 3rd-grade mathematics courses, providing question- and knowledge-concept-level sequences alongside auxiliary question texts, explanations, and images.

Dataset Link

Data Scale and Format

Metrics

Scale

Description

Student Count

18,066

14,453 in train/validation sets, 3,613 in independent test set

Question Count

7,652

Valid questions contained in metadata/questions.json

Knowledge Concept Path Count

1,175

Standard knowledge concept paths mapped in metadata/kc_routes_map.json

Question-Level Sequence Count

34,578

30,965 train/validation windows, 3,613 test set windows

Concept-Level Sequence Count

37,010

33,397 train/validation windows, 3,613 test set windows

Note

XES3G5M originates from the AAAI 2023 Global Knowledge Tracing Challenge; the official release is tailored for sequential modeling, pre-split into windows with maximum length 200. UniKT loads and evaluates on question-level sequences question_level by default.

Release Package Fields

The following fields are grouped by actual table headers in each CSV of the release package.

Base Columns Shared Across All Sequence Files

Field Name

Data Type

Description

fold

Integer

Fold index for train/val sets, values 0 to 4; -1 for test set

uid

Integer

Student ID

questions

String

Question ID sequence, comma-separated

concepts

String

Knowledge concept ID sequence, comma-separated; multiple concepts separated by underscores in question-level files

responses

String

Response correctness sequence, 1 for correct, 0 for incorrect

timestamps

String

Response start timestamp sequence, in milliseconds

Additional Columns by File

Field Name

Appearing Files

Description

selectmasks

Question-level train and window test files, kc_level/train_valid_sequences.csv, kc_level/test_question_window_sequences.csv

Batch valid bit mask; -1 indicates padding or historical-only interactions excluded from loss and evaluation

is_repeat

kc_level/train_valid_sequences.csv, kc_level/test.csv, kc_level/test_question_window_sequences.csv

1 indicates current concept belongs to the same question as preceding concept, 0 indicates a new question

cidxs

kc_level/test.csv

Interaction index sequence in concept-level test set

qidxs

kc_level/test_question_window_sequences.csv

Interaction index sequence in question-level test set

rest

kc_level/test_question_window_sequences.csv

Remaining knowledge concepts count for this interaction in original sequence, used for prediction fusion

orirow

kc_level/test_question_window_sequences.csv

Row index of this interaction in the test set

Question-level test file question_level/test_quelevel.csv contains base columns only, excluding selectmasks, is_repeat, cidxs, qidxs, rest, orirow.

Question Metadata (metadata/questions.json)

Field Name

Data Type

Description

content

String

Question stem text, preserving LaTeX

kc_routes

String

Knowledge concept path text

answer

String

Standard correct answer

options

List[String]

Multiple choice options, can be null for fill-in-the-blank questions

analysis

String

Explanation text, preserving LaTeX

type

Integer

Question type, 0 for fill-in-the-blank, 1 for multiple choice

metadata/kc_routes_map.json records mappings between knowledge concept IDs and syllabus path texts; metadata/embeddings/ and metadata/images/ provide vector representations and images for questions, explanations, and concepts respectively.