Bizagi Text Analytics connector, available for download at Bizagi Connector XChange, acts as a bridge between your Bizagi project and Microsoft Cognitive Services. With this connector you can easily use the Text Analytics API from Cognitive Services so you can identify and categorize important concepts, extract key phrases from unstructured texts, and improve the understanding of customer perception.
Identifying and categorizing concepts
This connector helps you classify a broad range of entities in a text, such as: people, places, organizations, dates, and percentages through entity recognition. You can also detect and extract more than 100 types of personally identifiable information (PII) and over 80 types of protected health information (PHI) in the analyzed texts. With this, you can enhance your processes with insights from prebuilt extraction models.
Extracting key phrases from unstructured text
You can identify rapidly the main ideas of an unstructured text. You will get a list of the relevant phrases that best describe each record's subject by relying on key phrase extraction. You can also pull and organize information to analyze important topic and trends.
Improving the understanding of customer perception
This connector helps you identify positive, neutral and negative sentiments in texts from social media, customer reviews and other sources. This will help you see the general customer concept of your brand.
How to use the Bizagi Text Analytics connector
The Bizagi Text Analytics connector sends a request to Microsoft Cognitive Services through the Text Analytics API. This means that you must have created an Azure resource for Text Analytics.
You need the service's key and endpoint to setup the connector in Bizagi Studio.
The following data model covers the basic structure for all the Cognitive Services responses.
The Text Analytics connector offers the following methods:
•Detect Language: returns the detected language with a numeric score that represents the certainty of the detected language.
•Named Entity Recognition: returns a list of named entities, such as: Person, Location and Organization and their values in the text.
•Linked Entities from a well-known knowledge base: returns a list of the named entities and their values along with links to knowledge bases.
•Key Phrases: returns a list of strings that highlight the key talking points of the input text.
•Sentiment: returns the input text's sentiment prediction along with the sentiment scores for each sentiment class: positive, negative and neutral.
For more detailed information about setting up the connector and its methods, check the connector's configuration documentation available for download in Bizagi Connector XChange.