Clarabridge’s Classification Suite organizes concepts and themes extracted from the underlying customer feedback data in a hierarchical structure for quick analysis and action. The NLP engine distills vast quantities of unstructured data to extract entities, events and their underlying structure using a variety of statistical and rules based algorithms. The Classification suite then automatically assigns a set of themes to the transformed data and organizes the themes in a hierarchy that is controlled by and is of importance to the user such as emerging issues, industry-based VOC concepts, business process based concepts, customer sentiment/emotion driven VOC concepts et al.
Easily Managed by Business Users
Clarabridge’s Classification Suite helps business users unlock and optimize the true business value of the concepts extracted from their customer feedback data. It hides the complexity of classification and instead presents an interactive interface for casual business users to understand the organization of the underlying data.
- It automatically classifies entire document sets into editable and logical categories without any need for manual effort or intervention. Users can discover and track re-occurring topics and themes without the need to enter a topic or search query.
- Business users are provided with the option to edit categories by changing, adding or deleting automatically generated categories before viewing by end-users. They can apply their domain knowledge towards fine tuning these categories to suit their business goals and needs.
- Clarabridge Classification Suite eliminates the rigors and efforts of maintaining categorization trees. As documents process, the Classification Suite organically grows and learns what categories “look like” and refines its performance accuracy by statistically and textually leveraging positive data instances as training sets.
- Categories maintain a high level of accuracy by automatically extracting relevant linguistic features from documents and associating them with their given categories. This advanced approach also maintains an optimum level of relevancy or coverage to ensure emerging concepts are not missed out.
Key Features
- Built-in categorization models for different industries and application areas
- Categorization Library with 400+ built-in categories that span a diverse set of Voice of Customer (VOC) themes
- Auto Categorization for automatic discovery of emerging themes and concepts
- Machine learning to discover hidden concepts and issues
- Ad Hoc search sand box for quick discovery of what is being discussed in the underlying data
- Wizard-based categorization management tools – cross reference, categorization model import/export, collaboration tools and much more