文本分析 API 的示例用户方案Example user scenarios for the Text Analytics API

文本分析 API 是一个基于云的服务,提供基于文本的高级自然语言处理。The Text Analytics API is a cloud-based service that provides advanced natural language processing over text. 本文将介绍有关将该 API 集成到业务解决方案和流程的一些示例用例。This article describes some example use cases for integrating the API into your business solutions and processes.

分析调查结果Analyze Survey results

使用情绪分析处理原始文本响应,从客户和员工调查结果中提取见解。Draw insights from customer and employee survey results by processing the raw text responses using Sentiment Analysis. 聚合分析结果,跟进趋势,并提升客户参与度。Aggregate the findings for analysis, follow up, and driving engagements.

描述如何对客户和员工调查执行情绪分析的插图。

分析录制的客户来电Analyze recorded inbound customer calls

使用文本转语音、情绪分析和关键短语提取从客户服务通话中提取见解。Extract insights from customer services calls using Text to Speech, Sentiment Analysis, and Key Phrase Extraction. 在 Power BI 仪表板或门户中显示结果,以便更好地了解客户,突出显示客户服务趋势,并提升客户参与度。Display the results in Power BI dashboard or a portal to better understand customers, highlight customer service trends, and drive customer engagement. 以批的形式发送 API 请求以生成报告,或者进行实时干预。Send API requests as a batch for reporting, or in real-time for intervention. 请参阅 GitHub 上的示例代码。See the sample code on GitHub.

描述如何使用情绪分析从客户服务通话中自动获取见解的插图

处理和分类支持事件Process and categorize support incidents

使用关键短语提取和实体识别来处理以非结构化文本格式提交的支持请求。Use Key Phrase Extraction and Entity Recognition to process support requests submitted in unstructured textual format. 使用提取的短语和实体将请求分类,以进行资源规划和趋势分析。Use the extracted phrases and entities to categorize the requests for resource planning and trend analysis.

描述如何使用关键短语提取和实体识别分类事件报告与趋势的插图

监视产品的社交媒体源Monitor your product's social media feeds

在产品的 Twitter 或 Facebook 页监视用户对产品提供的反馈。Monitor user product feedback on your product's twitter or Facebook page. 使用数据来分析客户对新推出产品的情绪,提取有关功能和功能请求的关键短语,或者在出现客户投诉时解决投诉。Use the data to analyze customer sentiment toward new products launches, extract key phrases about features and feature requests, or address customer complaints as they happen. 请参阅 Microsoft Flow 模板示例。See the example Microsoft Flow template.

描述如何使用关键短语提取在社交媒体上监视产品和公司反馈的插图

对包含敏感信息的文档进行分类和编修Classify and redact documents that have sensitive information

使用“命名实体识别”识别文档中的个人信息和敏感信息。Use Named Entity Recognition to identify personal and sensitive information in documents. 使用数据对文档进行分类或对其进行编修,以便安全地共享这些文档。Use the data to classify documents or redact them so they can be shared safely.

描述如何使用 NER 检测个人信息并对文档进行分类和编修的图像

进行观点挖掘Perform opinion mining

将调查、客户反馈或评论(其文本包含某方面的观点)中与产品或服务的特定方面相关的观点进行分组。Group opinions related to specific aspects of a product or service in surveys, customer feedback, or wherever text holds an opinion about an aspect. 使用它来帮助指导产品启动和改进、营销工作,或者重点介绍你的产品或服务的表现情况。Use it to help guide product launches and improvements, marketing efforts, or highlight how your product or service is performing.

有关酒店的示例观点。

后续步骤Next steps