NFT Bidding Behaviours and Sentiment Analysis

This folder serves as a repository for hosting the raw data and jupyter notebook files needed to recreate the analysis for 'Dutch auction dynamics in non-fungible token (NFT) markets', published in the Economic Modelling journal.

Purpose of File 'Telegram_Sentiment_Analysis_using_VADER.ipynb':
1. Deconstructing and cleaning 30,197 Telegram Forum messages to extract further insights on the bidding and listing behaviours of investors involved in a fledgling NFT marketplace
2. Applying a Sentiment Analysis model specialising in social media communications (VADER)
3. Extracting sentiment scores, for later analysis against NFT listing and sales data
4. Visually inspecting the evolution of forum participation and investor sentiment over an 8-week study period.

Purpose of File 'NFT_Plots.ipynb':
1. Visually inspecting the evolution of 28,919 NFT listings and 4,937 NFT sales data over an 8-week period.

The data required to run both above notebooks are available in the 'NFT_Dutch_Auction_Study_Data_Regressions.xlsx' file

Purpose of File 'NFT_Dutch_Auction_Study_Network_Graph_of_Wallets.ipynb':
1. Visually inspect the inter-connections between all wallets buying and selling on the NFT marketplace, where each wallet represents a node. This is one method of revealing suspicious trading patters amongst marketplace users.

The data required to run the above notebook is available in the NFT_Dutch_Auction_Study_Data_Network_Analysis.xlsx' file

Jupyter Notebook(s) Author: Dr. Darren Shannon. All errors my own.
Journal Article Authors: Dr. Darren Shannon, Prof. Michael Dowling, Dr. Marjan Zhaf, Dr. Barry Sheehan

More information on running the files:
- Minor code amendments are required in each notebook to open the data file and save generated images locally.
- Although minimial code adjustments are required to get the code working, it is suggested that you have a beginning level of proficiency in Python to run the code.
- Virtual environment / project generation / library / root files are not provided and will need to be created before running the Python code.
- Information on the library packages required to run the analysis are provided in the relevant notebooks.
- Please direct any detected errors in the code / data to: [e-mail]

本文件夹作为存储原始数据及用于重现《经济建模》期刊中‘非同质化代币（NFT）市场中的荷兰式拍卖动态’分析所需的 Jupyter Notebook 文件的库。

文件 'Telegram_Sentiment_Analysis_using_VADER.ipynb' 的用途如下：
1. 解构并清洗 30,197 条 Telegram 论坛消息，以进一步洞察参与新兴 NFT 市场的投资者在投标和挂牌方面的行为。
2. 应用专注于社交媒体通信的 sentiment analysis 模型（VADER）。
3. 提取情感得分，以备后续与 NFT 挂牌和销售数据进行对比分析。
4. 在为期 8 周的研究期间，对论坛参与度和投资者情绪的演变进行可视化分析。

文件 'NFT_Plots.ipynb' 的用途如下：
1. 在为期 8 周的时段内，对 28,919 条 NFT 挂牌和 4,937 条 NFT 销售数据进行可视化分析。

运行上述 Notebook 所需的数据存储在 'NFT_Dutch_Auction_Study_Data_Regressions.xlsx' 文件中。

文件 'NFT_Dutch_Auction_Study_Network_Graph_of_Wallets.ipynb' 的用途如下：
1. 视觉化检查在 NFT 市场进行买卖的所有钱包之间的相互连接，其中每个钱包代表一个节点。这是揭示市场用户之间可疑交易模式的一种方法。

运行上述 Notebook 所需的数据存储在 'NFT_Dutch_Auction_Study_Data_Network_Analysis.xlsx' 文件中。

Jupyter Notebook 作者：Darren Shannon 博士。所有错误均由本人承担。
期刊文章作者：Darren Shannon 博士、Michael Dowling 教授、Marjan Zhaf 博士、Barry Sheehan 博士

关于运行文件的更多信息：
- 每个 Notebook 中需要对代码进行细微调整，以打开数据文件并本地保存生成的图像。
- 尽管需要微小的代码调整才能使代码工作，但建议您具备基本的 Python 技能以运行代码。
- 虚拟环境、项目生成、库和根文件未提供，需要在运行 Python 代码之前创建。
- 运行分析所需的库包信息提供在相关 Notebook 中。
- 请将检测到的代码/数据错误反馈至：[e-mail]