CartoGAZE

CartoGAZE constitutes a large eye movement dataset produced during the observation of cartographic products. The production of this dataset was based on the collection of gaze data (SMI RED250) during the observation of the snapshots of 2D static Google navigational maps under free-viewing conditions. The dataset includes the map stimuli, the AOI files, the task descriptions, and full procedural details for the reproducibility of results and to create possibilities for future research.

We aim to develop an automatic and semi-automatic AOI-based analysis model capable of detecting attentive areas from the point of view of human operators, by considering the vector characteristics and visual variables of the map features, task difficulty, expertise, and spatial memory strategies of human operators.

A full description of the CartoGAZE dataset can be found in (please cite the work below as a reference to CartoGAZE):

Keskin, M., Krassanakis, V., & Coltekin, A. (2023). Visual Attention and Recognition Differences Based on Expertise in a Map Reading and Memorability Study, In (eds. Popelka, S., Stachoň, Z., Kiefer, P., Çöltekin, A.) Eye-Tracking in Cartography Special Issue, IJGI - MDPI. doi: 10.3390/ijgi12010021

https://www.mdpi.com/2220-9964/12/1/21

CartoGAZE is freely distributed to the scientific community via Harvard Dataverse.
The raw data was collected in the framework of the doctoral research project entitled “Exploring the Cognitive processes of Map Users Employing Eye Tracking and EEG”.

Next to the data_description file, there are six folders in CartoGAZEDatabase.zip; Raw_ET_data, AOIs, Map_stimuli, Final_Fixations, AOIs_all_results, and Calculate_AOI_areas folders; all of which are explained below:
1. Raw_ET_data: The collected eye tracking data in txt format, and can be linked with the map stimuli using the SYNC number in lines with, e.g., # Message: SYNC 103

2. AOIs: The coordinates of AOIs are shared as csv file format.

3. Map stimuli: The snapshots of 2D static Google navigational maps were taken at zoom level 15 and approx. 1:40k scale. The total number of map stimuli used in the experiment is 37 and can be found under Map_stimuli folder in png format. The resolutions of stimuli is provided in excel format in the same folder.

4. The outputs of the scripts can be found in Final Fixations and AOIs_all_results folders as single .csv files organized per stimulus, map feature type, participant and task type. Explanation of columns are as follows:
Column 1: Gaze Position X (Left eye)
Column 1: Gaze Position Y (Left eye)
Column 1: Trial Duration
Column 1: Duration Start
Column 1: Duration End
Column 1: The number of points clustered in the same fixation

In AOI files, the rest of the columns after the above-mentioned 6 columns represent within- or outside-AOIs:
- If the summary of the rest of the columns ≤1, the fixation is inside the AOI, else outside AOI (represented as 0s and 1s).
These files under AOIs_all_results are also aggregated in one .csv file named “AOI_all_results”.

5. Calculate_AOIs_areas: includes the areas of AOIs (square pixel) separately for each stimulus and map feature type, and aggregated in a single file named “all_AOI_areas.csv”.

CartoGAZE是一个大型眼动数据集，其数据采集于用户观察地图产品的过程。该数据集的构建基于自由观察条件下，用户观看2D静态谷歌导航地图快照时采集的眼动数据（SMI RED250）。该数据集包含地图刺激材料、兴趣区域（AOI）文件、任务描述，以及完整的流程细节——这些细节旨在确保结果可复现，并为未来研究创造可能。

我们旨在开发一种基于兴趣区域（AOI）的自动与半自动分析模型，该模型能够从人类操作者视角检测注意力区域——其核心在于考虑地图要素的矢量特征与视觉变量、任务难度、操作者专业水平及空间记忆策略。

关于CartoGAZE数据集的完整描述可参见（引用该数据集时请参考以下文献）：Keskin, M., Krassanakis, V., & Coltekin, A. (2023). 基于专业水平的地图阅读与记忆研究中的视觉注意力及识别差异，载于（编委会：Popelka, S., Stachoň, Z., Kiefer, P., Çöltekin, A.）《地图学中的眼动追踪》特刊，IJGI - MDPI。DOI：10.3390/ijgi12010021 https://www.mdpi.com/2220-9964/12/1/21

CartoGAZE可通过Harvard Dataverse向科学界免费获取。

在CartoGAZEDatabase.zip压缩包中，除data_description文件外，还包含六个文件夹：Raw_ET_data、AOIs、Map_stimuli、Final_Fixations、AOIs_all_results及Calculate_AOI_areas；各文件夹说明如下：

1. Raw_ET_data：该文件夹包含txt格式的眼动采集数据，可通过行中的SYNC编号（例如# Message: SYNC 103）与地图刺激材料关联。

2. AOIs：该文件夹以csv格式提供兴趣区域（AOI）的坐标数据。

3. Map_stimuli：该文件夹包含2D静态谷歌导航地图快照，拍摄于缩放级别15（约1:40000比例尺）。实验所用地图刺激材料共37张，以png格式存储于此文件夹；刺激材料的分辨率信息以excel格式文件提供在同一文件夹中。

4. Final_Fixations与AOIs_all_results：脚本输出文件以单个.csv格式存储于Final_Fixations与AOIs_all_results文件夹中，按刺激材料、地图要素类型、参与者及任务类型分类组织。列说明如下：列1：左眼凝视位置X坐标；列1：左眼凝视位置Y坐标；列1：试验时长；列1：时长起始；列1：时长结束；列1：同一凝视点中聚类的点数。在AOI文件中，上述6列之后的其余列表示凝视点位于兴趣区域（AOI）内或外：- 若其余列的总和≤1，则该凝视点位于AOI内，否则位于AOI外（以0和1表示）。AOIs_all_results文件夹下的文件还被汇总为一个名为"AOI_all_results"的.csv文件。

5. Calculate_AOI_areas：该文件夹包含各刺激材料及地图要素类型对应的兴趣区域（AOI）面积（单位：平方像素），并汇总为一个名为"all_AOI_areas.csv"的文件。