Advanced Carbon Footprint Prediction Using Hybrid Machine Learning and Ai-Assisted Recommendations
ハイブリッド機械学習とAI支援レコメンデーションを用いた高度なカーボンフットプリント予測 (AI 翻訳)
Inchara R, Madhu M. Nayak
🤖 gxceed AI 要約
日本語
本論文では、ユーザーの活動ログとIoTセンサー（ESP32）からのリアルタイムデータを統合したカーボンフットプリント予測・推奨システム「CarbonIQ」を提案する。ランダムフォレスト回帰モデルとセンサーフュージョン機構により炭素排出量を予測し、生成AIが個別最適化された削減推奨を提供する。実験結果から、予測精度と行動促進効果が確認された。
English
This paper introduces CarbonIQ, a system integrating IoT sensor data (ESP32) and user activity logs for carbon footprint prediction using a Random Forest regression model with hybrid sensor fusion. A generative AI module provides personalized carbon reduction recommendations based on user behavior and past actions. Experimental results demonstrate effective emissions prediction and actionable recommendations.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
本論文は個人レベルのカーボンフットプリント予測にAIを活用するもので、日本の企業における従業員のカーボンアクション促進や、家庭の排出量見える化に応用できる可能性がある。ただし、日本の開示規制（SSBJなど）との直接的な関連は薄い。
In the global GX context
This paper presents an AI-driven personal carbon footprint prediction and recommendation system. While it does not directly address corporate disclosure frameworks like TCFD or ISSB, it demonstrates how machine learning and generative AI can be applied to individual-level emission tracking and behavior change, which could complement broader sustainability initiatives.
👥 読者別の含意
🔬研究者:Researchers can examine the hybrid sensor-fusion and generative recommendation methodology for carbon footprint applications.
🏢実務担当者:The system could inspire corporate sustainability teams to develop internal tools for employee carbon footprint tracking and reduction.
📄 Abstract（原文）
Rising levels of carbon emissions have emerged as a key factor to climate change requiring smart mechanisms of monitoring and mitigation. In this paper, CarbonIQ, a machine learning-based, generative AI-based, and IoT-based data collection integrated carbon footprint prediction and recommendation system will be introduced. The system uses user activity logs and real-time sensor values received by an ESP32 which has been pre-configured with the sensor tags. Gassed to the carbon emission estimation and prediction is connected to gas sensors. The proposed approach uses the Random Forest regression model with a hybrid sensor-fusion mechanism. Besides prediction, a generative AI module allows giving personalized carbon reduction recommendations based on users' actions and their past actions. This system has also been optimized with ranking-based feedback, which will increase the engagement with the system and encourage users to move forward. sustainable practices. That the proposed strategy is useful can be validated with experimental results that demonstrate important outcomes in emissions prediction and taking actionable steps. Incorporating prediction and recommendation into a single structure, one would be able to make better decision. and develop more data- driven solutions to sustainability. Rising levels of carbon emissions have emerged as a key factor to climate change requiring smart mechanisms of monitoring and mitigation. In this paper, CarbonIQ, a machine learning-based, generative AI-based, and IoT-based data collection integrated carbon footprint prediction and recommendation system will be introduced. The system uses user activity logs and real-time sensor values received by an ESP32 which has been pre-configured with the sensor tags. Gassed to the carbon emission estimation and prediction is connected to gas sensors. The proposed approach uses the Random Forest regression model with a hybrid sensor-fusion mechanism. Besides prediction, a generative AI module allows giving personalized carbon reduction recommendations based on users' actions and their past actions. This system has also been optimized with ranking-based feedback, which will increase the engagement with the system and encourage users to move forward. sustainable practices. That the proposed strategy is useful can be validated with experimental results that demonstrate important outcomes in emissions prediction and taking actionable steps. Incorporating prediction and recommendation into a single structure, one would be able to make better decision. and develop more data- driven solutions to sustainability.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.47392/irjaeh.2026.0601first seen 2026-07-13 06:14:53
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。