UiPath Cognitive Activities

The Cognitive Activities package helps you use Google's, IBM's, Stanford's and Microsoft's APIs, and automatically process the information that they help you extract.

Overview of the Cognitive Activities Package

The Cognitive Activities package facilitates the integration of leading APIs from Google, IBM, Stanford, and Microsoft into your workflows, streamlining the automated processing of extracted information. This comprehensive package empowers users to seamlessly translate text between languages and extract pertinent details from provided text, including overall sentiment, key phrases, potential errors, and the detected language.

API Key Requirement

It is essential to note that the utilization of all cognitive activities within workflows necessitates a valid API key. Regional keys are applicable to StanfordNLP Text Analysis, Google Text Analysis, IBM Watson NLU Text Analysis, and Microsoft Text Analysis.

Activities in the Cognitive Activities Package

The Cognitive Activities package contains the following activities:

• Google Text Analysis - Extracts the language of a specified text, and the strength, positivity or negativity of the sentiment.
• Google Text Translate - Translates a specified text from one selected language to another, and extracts the result as a String variable.
• IBM Watson NLU Text Analysis - Extracts key information from a specified text, and returns multiple IBM Watson NLU specific data types as well as a raw result in .json format, stored in a String variable.
• IBM Watson Text Analysis - Extracts the language of a specified text, and an analysis of it in a JSON format, as a String variable.
• Microsoft Text Analysis - Extracts the language of a specified text, the overall sentiment, key phrases as IEnumerable(String) variables, and the entire information, in a JSON format.
• StanfordNLP Text Analysis - Analyzes a specified text by using the Stanford CoreNLP Natural Language Annotation Technology, by making API calls to the Stanford CoreNLP server. There are two types of output for this activity, the Raw Result field outputs a .json string that you can further parse, while the Sentences field outputs a list of StanfordCoreNlpSentence variables that can be further processed by using other Stanford CoreNLP activities.
• StanfordNLP Get Components - Processes a sentence stored in an StanfordCoreNlpSentence variable, retrieved from a Text Analysis activity, and returns a list of Tuple variables formed from each word and its corresponding analyzed element.
• StanfordNLP Get OpenIE Relations - Extracts the OpenIE relations of a StanfordCoreNlpSentence variable retrieved from a Text Analysis activity and stores them in a Tuple variable. The strings in this variable represent a subject, a relation and the object of the relation in this order.
• StanfordNLP Get Sentence Sentiment - Extracts the sentiment and sentiment value from a StanfordCoreNlpSentence variable retrieved from a Text Analysis activity and returns them in a Tuple variable. The first string in the Tuple represents the sentiment, while the second represents the sentiment value.

Features and Applications of the Cognitive Activities Package

The Cognitive Activities are a versatile package for various use cases, leveraging the capabilities of Google, IBM, Stanford, and Microsoft APIs:

• Multilingual Content Management
• Automated Document Summarization
• Language Detection and Categorization
• Advanced Linguistic Analysis