ИИ-классификация тикетов поддержки электронной коммерции

Дом

Список AI-промптов

Классификация тикетов электронной коммерции

${:__('collect %s',Классификация тикетов электронной коммерции)}$

Используйте обработку естественного языка для анализа тикетов поддержки клиентов, определения намерения и тональности, а также автоматического присвоения предопределенных бизнес-категорий для повышения эффективности ответа и точности данных.

Содержание промпта

Копировать

Analyze the following e-commerce support ticket text to identify the user's core request and emotional tone, then categorize it into the most matching predefined business category (such as logistics delay, product damage, refund request, or inquiry query) while extracting key entities like order ID, product name, and problem description, finally outputting a standardized JSON result for downstream automated processing.

Копировать

Проанализируйте следующий текст тикета поддержки электронной коммерции, чтобы определить основную просьбу пользователя и эмоциональный тон, затем классифицируйте его в наиболее соответствующую предопределенную бизнес-категорию (например, задержка доставки, повреждение товара, запрос на возврат средств или запрос информации), одновременно извлекая ключевые сущности, такие как номер заказа, название продукта и описание проблемы, и в конечном итоге выведите результат в стандартизированном формате JSON для последующих процессов автоматизированной обработки.

Копировать

Поделиться Нажмите, чтобы скопировать ссылку

Комментарии (0)

0/300

Представлять на рассмотрение

Рекомендация

Классификатор тикетов E-commerce

Act as an intelligent customer service system for an e-commerce platform. You will receive raw user inquiry texts and must analyze their core intent to classify them into one of five predefined standard categories: Logistics Inquiry, Return/Exchange Request, Product Consultation, Complaint/Suggestion, or Account Issue. Carefully review the input, extract key entities such as order numbers, product names, or specific pain points, and ignore irrelevant pleasantries. Finally, output only a standardized JSON result containing the original text, the identified intent label, and a confidence score, ensuring accurate classification to support subsequent automated processing workflows.

Автоответ CS E-commerce

Act as a senior ecommerce customer service expert. Generate a professional, friendly, and accurate response based on the provided customer inquiry and product knowledge base. The reply must directly address the customer's issue, avoid mechanical template language, and maintain brand tone consistency to ensure the customer feels valued and understood.

Ответ службы поддержки EC

Act as a senior e-commerce customer service expert. Based on the provided product details (including name, price, specifications, stock status, and key selling points) and the user's specific inquiry, generate a natural, friendly, and professional reply text. The reply should directly address the user's concerns, accurately cite key product information, avoid mechanical template language or overly complex jargon, and ensure the tone aligns with brand standards to facilitate conversion.

Ответ службы поддержки

Act as a senior e-commerce customer service expert. Based on the provided order ID, current logistics status, and user's emotional keywords, generate a professional yet empathetic reply. The response must include a sincere apology for the delay, a specific estimated delivery time, and a small coupon as compensation. Maintain a friendly tone that solves the problem while avoiding robotic template language.

Тикет поддержки электронной коммерции

Act as an intelligent customer service assistant for an e-commerce platform. Analyze the user's after-sales inquiry text to identify core intents such as returns, exchanges, or logistics queries. Based on the current order status and platform policies, generate a friendly and professional reply draft that directly addresses the user's concerns and guides them to the next steps.

Классификация тикетов поддержки

Act as part of an e-commerce customer service system. Analyze the user's raw inquiry text and classify it into one of four categories: pre-sales consultation, after-sales service, logistics query, or complaint/suggestion based on its semantic intent. Output the corresponding category label and a brief justification to ensure accurate classification that assists human agents in handling tickets quickly.