DVI

Overview

DVI (Data Visualization Interface) is a powerful tool for managing data and visualizing hashtag networks. It integrates with multiple related projects to streamline data mining, analysis, and visualization workflows. A core feature is the creation of mining tasks, each assigned a unique Mining Unique Identifier (MUID) that links tasks across all integrated tools.

---

Features

• CRUD Operations: Perform Create, Read, Update, and Delete operations seamlessly.
• Mining Tasks: Generate tasks that integrate with:
  • IDMB: Scrapes hashtag conversations.
  • Graffiti Detection OD TensorFlow: Detects graffiti in image datasets.
  • Hashtag Custom NER spaCy: Analyzes mined hashtags.
• MUID (Mining Unique Identifier): Links tasks across projects.
• Graph Visualization: Offers a dynamic interface for analyzing hashtag networks.
• Customizable Visualization: Modify graph parameters and filters to suit specific needs.
• User-Friendly Interface: Intuitive design for efficient task management.

---

Project Structure

dvi/
├── database/          # Database-related files and configurations
├── db_actions/        # Scripts for database operations
├── docs/              # Documentation and resources
├── includes/          # Shared includes and utilities
├── json/              # JSON-related data and configurations
├── json_actions/      # Scripts for JSON data manipulation
├── block_data_live.php
├── db.php
├── delete_task.php
├── edit.php
├── functions.php
├── hashtags.php
├── hashtags_ai_data.php
├── hashtags_ai_data_live.php
├── header.php
├── index.php
├── json_data.php
├── json_hashtag.php
├── json_scandir.php
├── json_scandir_h.php
├── json_user.php
├── save_task.php
├── sigma.php

Key Components

• hashtags_ai_data_live.php: Generates a JSON cache file representing hashtag network structures.
  • Supports graph visualization workflows.
  • Enables cleaning messy hashtag records dynamically.
  • Visualization parameters can be adjusted via form inputs or left as defaults.

---

MUID Details

The MUID (Mining Unique Identifier) is a key feature of DVI. It is a compound identifier automatically generated for each mining task and serves as a transversal key across all related projects.

MUID Structure

1. Seed Node: The main hashtag or entry point.
   Example: afeks.

2. Mining Depth: Depth of the mining process (0 to 4).
   Example: 1.

3. Mining Type: Task type, such as:

   • hashtagTop: Extracts the most valuable posts for a hashtag.
     Example: hashtagTop.

4. Hashtag Media Amount: Total number of media posts mined.
   Example: 9.

5. Unique Hash: A short, randomly generated hash ensuring uniqueness.

Example MUID

afeks_1_hashtagTop_9_7052518d

---

Graph Visualization

All graph visualization features are managed in hashtags_ai_data_live.php.

The file is commented for clarity, and the visualization parameters can be modified as needed. Users can either provide custom values through the interface or rely on the defaults.

Key Parameters

• Graph Layout and Physics:

  • initialLayout: Default is circlepack.
  • autoGravityScale: Default is "auto".
  • nodeMinDegree: Default is 0.
  • gravity: Default is 1.
  • iterations: Default is 133.
  • scale: Default is 5000.

• Node and Edge Behavior:

  • adjustSizes: Default is 'false'.
  • cleanEntities: Default is 'true'.
  • barnesHutOptimize: Default is 'false'.
  • barnesHutTheta: Default is 0.5.

• Network Filters:

  • networkfilter: Default includes "standard", "text_ai", "image_ai", and "text_ai_entities".
    Example:

    $networkfilter_get = isset($_GET['networkfilter']) ? $_GET['networkfilter'] : ["standard", "text_ai", "image_ai", "text_ai_entities"];

Example Code Snippet

$initialLayout = isset($_GET['initialLayout']) ? $_GET['initialLayout'] : "circlepack";
$nodeMinDegree = isset($_GET['nodeMinDegree']) ? $_GET['nodeMinDegree'] : 0;
$iterations = isset($_GET['iterations']) ? $_GET['iterations'] : 133;

---

Installation

1. Download the repository:
   Clone or download the ZIP file from the repository.

2. Upload files to your server:
   Extract and upload to your desired FTP folder.

3. Configure the database:
   Update db.php with your database credentials:

   <?php
   session_start();
   
   $conn = mysqli_connect(
     '[SERVER]',
     '[DATABASE_USER]',
     '[DATABASE_PASSWORD]',
     '[DATABASE_NAME]'
   ) or die(mysqli_error($mysqli));
   ?>

4. Access the application:
   Navigate to the uploaded folder in your browser:

   http://your-server-path/dvi

---

License

This project is licensed under the MIT License.

---

Feel free to suggest edits or report issues in the Issues section!