Skip to content

Sherbieny/chatbotto_lumen

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

47 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Chatbotto Lumen Application

This project is a rule-based chatbot application built with Nextjs and Lumen. It uses pgroonga for full-text search and mecab for Japanese language tokenization. The chatbot produces suggestions based on the user's input and answers questions directly from the qa table data.

Alternative versions

Features

  • Rule-based chatbot
  • Full-text search
  • Japanese language tokenization
  • Dockerized development environment
  • Nextjs frontend and admin dashboard for managing the chatbot settings and uploading the qa table data
  • Lumen backend API
  • Lumen console commands to import and index qa table data
  • CRON job to execute the lumen console command when triggered from the admin dashboard
  • Sample qa table data json files for import extracted from JaQuad dataset
  • Images and icons are generated by ChatGPT DALL-E

Installation

  1. Clone the repository

  2. add XDEBUG_CLIENT_HOST to your os environment variables

    a. For Linux or WSL2, add the following to your .bashrc or .zshrc vim ~/.bashrc or vim ~/.zshrc

    export XDEBUG_CLIENT_HOST=$(ip addr show docker0 | grep -Po 'inet \K[\d.]+')

    b. For Windows, Mac with Docker Desktop, manually add host.docker.internal to the docker-compose.yml file in the args: section of the back service

    args:
      XDEBUG_CLIENT_HOST: host.docker.internal
  3. Run the following commands in project root to build and start the docker containers

docker-compose up -d --build
  1. SSH into the back container (php-fpm/Debian) and run the following commands to install the dependencies

    a. Install composer dependencies

    composer install

    b. Migration

    php artisan migrate

    c. Import qa table data (optional)

    php artisan import:qa-data

    d. Index the qa table data (optional)

    php artisan index:qa-data

Configuration

  1. Update the docker-compose.yml back service environment variables

    a. APP_ENV - set to local for development and production for production b. APP_URL - set to http://localhost:8000 for development and https://yourdomain.com for production c. APP_DEBUG - set to true for development and false for production

  2. In front directory rename the .env.local.tmp to .env.local and update the following environment variables

    a. NEXT_PUBLIC_API_URL - set to http://localhost:8000 for development and https://yourdomain.com for production

  3. (optional) update dockerfiles/back/cronjob to set the CRON job schedule

    a. The default schedule is set to run every 5 minutes

API Endpoints

# Get suggestions based on the user's input
GET /api/suggestions
# Parameters: query (string)

# Send a message to the chatbot
POST /api/message
# Parameters: query (string)

# Get the chatbot settings
GET /api/settings

# Update the chatbot settings
POST /api/settings
# Parameters: settings (object)

# Upload the qa table data
POST /api/upload
# Parameters: file (file)

# Process the qa table data (import and index)
POST /api/process
  • Process the qa table data (import and index)

Usage

  1. Start the docker containers
docker-compose start
  1. Open the browser and navigate to http://localhost:3000 to access the chatbot application

  2. start typing to interact with the chatbot, select from the suggestions or ask a question directly

  3. Click on the gear icon to access the admin dashboard

  4. Use the admin dashboard to manage the chatbot settings and upload the qa table data a. On qa data upload, the CRON job will execute the lumen console command to import and index the qa table data. The CRON job is set to run every 5 minutes.

Tools and Technologies

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

GLP-3.0


チャットボット Lumen アプリケーション

このプロジェクトは、Nextjs と Lumen を使用して構築されたルールベースのチャットボットアプリケーションです。pgroonga を使用して全文検索を行い、mecab を使用して日本語のトークン化を行います。チャットボットは、ユーザーの入力に基づいて提案を生成し、qa テーブルデータから直接質問に答えます。

代替バージョン

特徴

  • ルールベースのチャットボット
  • 全文検索
  • 日本語のトークン化
  • Docker 化された開発環境
  • Nextjs フロントエンドと管理ダッシュボード
  • Lumen バックエンド API
  • Lumen コンソールコマンドを使用して qa テーブルデータをインポートおよびインデックスするための Lumen コンソールコマンド
  • 管理ダッシュボードからトリガーされた場合に Lumen コンソールコマンドを実行する CRON ジョブ
  • JaQuad データセットから抽出されたインポート用のサンプル qa テーブルデータ json ファイル
  • ChatGPT DALL-E によって生成された画像とアイコン

インストール

  1. リポジトリをクローンします

  2. OS 環境変数に XDEBUG_CLIENT_HOST を追加します

    a. Linux または WSL2 の場合、次のように .bashrc または .zshrc に追加します vim ~/.bashrc または vim ~/.zshrc

    export XDEBUG_CLIENT_HOST=$(ip addr show docker0 | grep -Po 'inet \K[\d.]+')

    b. Windows、Docker Desktop を使用している Mac の場合、back サービスの docker-compose.yml ファイルの args: セクションに手動で host.docker.internal を追加します

    args:
      XDEBUG_CLIENT_HOST: host.docker.internal
  3. プロジェクトのルートで次のコマンドを実行して Docker コンテナをビルドおよび起動します

    docker-compose up -d --build
  4. back コンテナ(php-fpm/Debian)に SSH して、次のコマンドを実行して依存関係をインストールします

    a. Composer 依存関係をインストールします

    composer install

    b. マイグレーション

    php artisan migrate

    c. qa テーブルデータをインポート(オプション)

    php artisan import:qa-data

    d. qa テーブルデータをインデックス化(オプション)

    php artisan index:qa-data

設定

  1. docker-compose.yml の back サービスの環境変数を更新します

    a. APP_ENV - 開発環境の場合は local に設定し、本番環境の場合は production に設定します b. APP_URL - 開発環境の場合は http://localhost:8000 に設定し、本番環境の場合は https://yourdomain.com に設定します c. APP_DEBUG - 開発環境の場合は true に設定し、本番環境の場合は false に設定します

  2. front ディレクトリで .env.local.tmp を .env.local にリネームし、次の環境変数を更新します

    a. NEXT_PUBLIC_API_URL - 開発環境の場合は http://localhost:8000 に設定し、本番環境の場合は https://yourdomain.com に設定します

  3. (オプション) dockerfiles/back/cronjob を更新して CRON ジョブのスケジュールを設定します

    a. デフォルトのスケジュールは 5 分ごとに実行されるように設定されています

API エンドポイント

# ユーザーの入力に基づいて提案を取得します
GET /api/suggestions
# パラメータ: query (string)

# チャットボットにメッセージを送信します
POST /api/message
# パラメータ: query (string)

# チャットボットの設定を取得します
GET /api/settings

# チャットボットの設定を更新します
POST /api/settings
# パラメータ: settings (object)

# qa テーブルデータをアップロードします
POST /api/upload
# パラメータ: file (file)

# qa テーブルデータを処理します(インポートおよびインデックス)
POST /api/process

使い方

  1. Docker コンテナを起動します
docker-compose start
  1. ブラウザを開いて、http://localhost:3000 に移動して、チャットボットアプリケーションにアクセスします

  2. タイプを開始して、チャットボットと対話し、提案から選択するか、直接質問をするかを選択します

  3. 歯車アイコンをクリックして、管理ダッシュボードにアクセスします

  4. 管理ダッシュボードを使用して、チャットボットの設定を管理し、qa テーブルデータをアップロードします a. qa データのアップロード時に、CRON ジョブは lumen コンソールコマンドを実行して qa テーブルデータをインポートおよびインデックスします。CRON ジョブは 5 分ごとに実行されるように設定されています。

ツールとテクノロジー

貢献

プルリクエストは歓迎です。重大な変更を行う場合は、まず問題を開いて議論してください。

ライセンス

GLP-3.0