Delete README.md

README.md

@@ -1,70 +0,0 @@
----
-license: apache-2.0
-task_categories:
-- text-retrieval
-- text-classification
-- tabular-classification
-language:
-- zh
-- ja
-- en
-tags:
-- nsfw
-- manga
-- doujinshi
-- metadata
-- recommendation-system
----
-
-# 🔞 Manga Gallery Metadata Archive (NSFW / 18+)
-
-## ⚠️ EXPLICIT CONTENT WARNING / 警告说明 ⚠️
-
-**[EN]** This dataset contains highly explicit **Not Safe For Work (NSFW)** and **18+** metadata related to adult manga and doujinshi. The text, tags, and titles included in this dataset describe sexually explicit acts, adult themes, and fetishes. **Do not proceed or download if you are under the age of 18, or if it is illegal to view such content in your jurisdiction.**
-
-**[ZH]** **18+ 强烈警告**：本数据集包含高度成人向（NSFW）的漫画与同人志**元数据（Metadata）**。数据集内的标题、标签、作者等文本信息涉及强烈的性暗示、成人主题及相关特殊偏好（XP）。**未满 18 岁者请立即关闭此页面。请确保下载和使用此数据符合您所在国家或地区的法律法规。**
-
----
-
-## Dataset Summary
-
-This dataset serves as a metadata archive designed to support local inventory retrieval and recommendation systems for adult manga and doujinshi.
-
-**Please note: This dataset contains ONLY text and structured metadata. It does NOT include any actual manga image files.**
-
-The data is primarily intended for the following natural language processing and data mining tasks:
-- **NLP Tag Semantic Aggregation:** Researching the distribution and synonym mapping of adult tags across different languages (Chinese/Japanese/English).
-- **Recommendation System Construction:** Developing personalized preference scoring algorithms based on multi-dimensional features (author, tags, keywords).
-- **RAG & Vector Retrieval:** Serving as an external knowledge base for Large Language Model (LLM) Retrieval-Augmented Generation.
-
-## Dataset Structure
-
-This repository contains two main data formats to facilitate various use cases:
-
-### 1. Structured CSV Files (`gallery_info/`)
-Contains all manga metadata collected from online sources or organized/cleaned locally.
-* **File Format:** `.csv`
-* **Use Cases:** Quick Pandas data analysis, NLP text preprocessing, or lightweight recommendation algorithm development.
-
-### 2. Database Dump (`dump-manga_gallery-202604271533.sql`)
-A complete MySQL database export file, preserving table structures such as primary keys and unique indices.
-* **File Format:** `.sql`
-* **Use Cases:** One-click restoration of a local manga database environment, ready for direct integration with SQL-query-based or ORM-backed application backends.
-
-## Data Fields
-
-Whether in CSV or SQL tables, the core data structure revolves around the following key fields:
-
-* `ID` (String): Unique identifier (e.g., gallery codes like `NH123456`, `JM123456`), serving as the database primary key and vector database ID.
-* `Title` (String): The full title of the manga/doujinshi, often including multi-language or translation group information.
-* `Tags` (String/List): A collection of related classification tags, including characters, scenes, and attributes (the core features for building a fetish/preference recommendation system).
-* `Author` (String): The name of the artist or author.
-* `Team/Group` (String): The name of the translation team or circle.
-* `Upload_Date` (Date): The original upload or publication date from the source.
-* *(Other potential extension fields such as page count, language, etc., depending on the specific scraping status).*
-
-## Usage Restrictions & Disclaimer
-
-1. **Research and Personal Use:** This dataset is intended strictly for technical research purposes (e.g., building a local gallery, testing vector retrieval models, LLM RAG experiments).
-2. **Disclaimer:** The uploader is not responsible for the tag or title text generated by third parties within the dataset. Users assume all legal and ethical responsibilities arising from the application of this data.
-3. **License Compliance:** Under the Apache License 2.0, if you distribute a modified version of this dataset, you must include prominent notices stating that you changed the files, and retain all original copyright, patent, trademark, and attribution notices.