AI-based enhancements include emotional and behavioral traits and a writeprint extension performing a stylometric analysis of documents
The new expert.ai NL API capabilities for emotional and behavioral traits debut with the “Sentiment & Opinion Mining Natural Language API” hackathon, running from May 6 to June 22, 2021
Expert.ai today announced advanced features enhancing analysis capabilities through its cloud-based natural language (NL) API. The new extension addresses one of the biggest challenges artificial intelligence developers face in the NL ecosystem – extracting emotions in large-scale texts and identifying stylometric data driving a complete fingerprint of content.
The ability to leverage the analysis of emotional and behavioral traits plays a critical role in advancing the development of intelligent NL-aware apps. By achieving insight into state of mind, developers and data scientists can go beyond the traditional positive or negative ratings of a sentiment analysis to identify the emotions of a text, enabling organizations to create applications that unlock the full potential of their data.
“From apps that analyze customer interactions, product reviews, emails or chatbot conversations, to content enrichment that increases text analytics accuracy, adding emotional and behavioral traits provides critical information that has significant impact,” said Luisa Herrmann-Nowosielski, Head of Product Management at expert.ai. “By incorporating this exclusive layer of human-like language understanding and a powerful writeprint extension for authorship analysis into our NL API, we are conquering a new frontier in the artificial intelligence API ecosystem, providing developers and data scientists with unique out-of-the-box information to supercharge their innovative apps.”
Emotional and behavioral traits
The expert.ai NL API captures a range of 117 different traits, providing the richest emotional and behavioral taxonomy available on the AI API market today. Emotional Traits are categorized into 8 different groups (anger, fear, disgust, sadness, happiness, joy, nostalgia, shame…). Behavioral Traits are divided into 7 groups (sociality, action, openness, consciousness, ethics, indulgence and capability) and the API assigns 3 levels of polarity (low, fair, high) to further indicate the level of each trait extracted.
The emotions and traits extension can be particularly useful to make media content categorization more effective and improve customer interactions by capturing new needs or advancing analytics by providing more detailed forecasting and enabling more effective recommendation tailoring for e-commerce and online advertising.
The expert.ai NL API writeprint extension performs a deep linguistic style analysis (or stylometric analysis) ranging from document readability and vocabulary richness to verb types and tenses, registers, sentence structure and grammar. Compare multiple documents to identify unique writing style and author invariants to streamline authorship analysis, establish the author of a specific text or isolate characteristics such as education level and other cultural aspects.
Writeprint features are also valuable for intelligence activities to establish, with a certain degree of certainty, whether a specific person wrote a text and identify forgeries or impersonations. They are also useful for publishing and media companies to categorize news and articles according to distinct writing styles (based on syntactic, structural and lexical features) or the readability level (that can be affected by slang, incorrect grammar etc.)
Sentiment & Opinion Mining NL API Hackathon
Today, expert.ai is launching a new “Sentiment & Opinion Mining Natural Language API” hackathon offering the chance to build an AI-based app leveraging the sentiment analysis, emotional and behavioral traits features provided by the NL API.
There is no entry fee for the hackathon. Individual developers and teams can subscribe at https://expertai-nlapi-042021.devpost.com until June 22, 2021.
Winners will be announced on July 15, 2021, with prizes worth a total of $10,000, including: $5,000 for 1st place; $2,500 for second place; $1,000 for 3rd place; and three $500 category prizes.
For more such updates and perspectives around Digital Innovation, IoT, Data Infrastructure, AI & Cybsercurity, go to AI-Techpark.com.