AI活用ECカスタマーサポート自動分類システム

家

AIプロンプト一覧

ECカスタマーサポートチケット分類

${:__('collect %s',ECカスタマーサポートチケット分類)}$

自然言語処理技術を活用し、ECプラットフォームのユーザー相談および苦情チケットを分析し、アフターセールス、物流、製品相談などの具体的な業務モジュールに自動分類することで、カスタマーサポートの応答効率と精度を向上させます。

プロンプト内容

コピー

Analyze the provided e-commerce customer consultation text to identify the core request and determine its business category. If the user mentions damaged, missing, or malfunctioning items, categorize it as an after-sales issue. If it involves delayed shipping, logistics updates, or lost packages, categorize it as a logistics issue. If it asks about product specifications, usage tutorials, or compatibility, categorize it as a product inquiry. Ignore emotional expressions and extract key facts for precise classification, outputting the corresponding standard business label.

コピー

提供されたECユーザー相談テキストを分析し、核心となる要求を特定してビジネスカテゴリを判断してください。商品破损、欠品、機能故障について言及している場合はアフターセールス問題、発送遅延、物流情報更新、宅配便紛失に関連する場合は物流問題、製品仕様、使用説明書、互換性について質問している場合は製品相談として分類してください。感情的な表現は無視し、重要な事実のみを抽出して正確に分類し、対応する標準ビジネスラベルを出力してください。

コピー

共有リンクをコピーするにはクリック

コメント (0)

0/300

送信

おすすめ

ECチケット分類

Please analyze the following e-commerce customer service ticket content, classify it into one of four categories: 'Logistics Inquiry', 'Return/Exchange Request', 'Product Quality Complaint', or 'Other Inquiry' based on business semantics, and briefly explain the reasoning. If the ticket involves multiple intents, prioritize the most urgent or core issue. The output format must include the classification label and a short reason, ensuring the result is accurate and aligns with actual business scenarios.

ECカスタマーサポートチケット分類

Analyze the following e-commerce support ticket text to identify the core issue type (such as logistics delay, product damage, refund dispute, or general inquiry), categorize it into the corresponding handling department based on predefined rules, extract key entities like order ID, product name, and user sentiment, and finally output a structured classification result with suggested handling solutions, ensuring accuracy and compliance with business standards.

ECチケット分類

Act as an intelligent customer service assistant for an e-commerce platform. Receive and analyze the user's raw inquiry text. Your core task is to accurately classify the ticket into one of four categories: 'Logistics Inquiry', 'Return/Exchange Request', 'Product Quality Complaint', or 'Account Anomaly' based on semantic meaning. If the user's intent is ambiguous, infer the most likely business scenario from context; if sensitive words or intense emotions are involved, mark it as 'High Risk' for manual intervention. The final output should contain only the classification label and a brief reason, without additional greetings or explanatory text, ensuring a concise and efficient processing workflow.

ECカスタマーサポートチケット分類

Analyze the following e-commerce customer service ticket content, identify the user's core issue, emotional tone, and specific request, and automatically categorize it into predefined business categories such as logistics inquiry, product quality issue, return or exchange application, or price dispute. Output the classification result along with a brief explanation of the reasoning to facilitate quick response and handling by the customer service team.

ECサポートチケット分類

Analyze the following e-commerce after-sales customer inquiry text and determine its business category based on semantics, including logistics inquiry, return/exchange application, product quality complaint, invoice issuance, and others. Output only the classification label without explaining the reasoning process, ensuring accuracy and compliance with business standards.

ECチケット分類

Analyze the following e-commerce customer inquiry ticket content, automatically classify it into predefined business categories such as logistics inquiry, return or exchange application, product consultation, or complaint suggestion, extract key entity information including order number, product name, and user sentiment, and finally output a structured classification result for the customer service team to handle quickly.