OBJECTIVES:

• Identify emotional tone of patient feedback and track issues that are documented in unstructured text
• Measure the effectiveness of AI interventions and its contribution to policy changes
• Monitor patient satisfaction and provide actionable insights for improving care delivery

METHODS:

-Gathering large amount of patient feedback from various sources such as .

-The application of NLP and machine learning techniques on large data from social media/review websites/survey responses to identify the sentiment of each feedback.

- AI solutions used are Geomapping, Sentiment algorithm, and contextual analytics to generate business ready insights with text visualization/graphs/tables

- Project outcome supports customer happiness teams at EHS to be able to analyze the feedback data using smart solutions and Artificial intelligence

RESULTS:

EHS has developed an innovative solution for accurately extracting people's opinions of unstructured texts and classifying them into sentiment classes using EHS Intelligence. The output of this project is a actionable report that can be used by healthcare providers to identify areas for improvement and tailor their services to better meet patient needs. Sentiment Analysis can also be integrated with other forms of data (demographic) to gain a more comprehensive understanding of patients' views.

• Improves the productivity of employees by a significant reduction of time to analyze, prioritize and faster action on feedbacks
• Improved customer service
• Feedback-based operational excellence and empowered patient community.

CONCLUSIONS:

More than 70% of feedback is Subjective Content and hence a smart algorithm was needed to convert the feedback into meaningful insights. Using machine learning and natural language processing, sentiment analysis algorithms were developed to automatically classify patient feedback into positive/negative/neutral categories, and identify specific themes and topics that are mentioned across multiple pieces of feedback. This can allow EHS to quickly identify areas of concern or areas where patients are particularly satisfied, and can help them make data-driven decisions about how to improve services.