X Homepage Platform About Us Consultancy Solutions References Blog Contact Log-in


Insight Natural Language processing - Text Analysis

In order to improve the customer experience by taking correct and rapid action from customer feedback, it is often not enough to only depend on structural survey scores.

Your customers offer valuable insights to your company, both through open-ended comments in feedback surveys and through the comments they share on social media and other online sites. However, these data are composed of unstructured textual expressions and are difficult to analyze at large scales.

Cloud4Feed Text Analysis solution, with its machine learning models developed with natural language processing techniques, conceptually correctly classifies the textual expressions of customers in an industry-specific manner, allowing unmet needs to be easily identified and maximum insights obtained from feedback.

Sentiment Analysis
Accurately categorizing the mood in the comments is very important for obtaining important insights from customer comments and taking the right actions.Textual expressions are separated into sentences, and each sentence is classified as positive / negative / neutral, and the sentiment score of each sentence is calculated separately.
It provides important insights by analyzing the topic detection model and numerical scores together.

Intent Analysis
The intention of each statement in customer comments in the Cloud4Feed Intent Analysis model is classified as Complaint, Acknowledgement, Suggestion, Abandoning and Information Request. It is possible to generate instant alarms and determine feedback actions according to these classification results.
While a customer with a tendency to Abandon in their comments requires very fast action, the opinions of customers who make Suggestions should be prioritized sensitively by including the relevant departments in the process.

Topic Detection
The Cloud4Feed supervised Topic Detection model enables the classification of textual expressions into industry-specific topics.
Thus, in addition to the determination of mood and intention distribution based on the topic, analysis such as Impact Analysis2 and NPS Driver Analysis are provided by combining them with numerical scores, from which important insights can be obtained.