[Product Review] Three Reasons Why We Use Cloud Natural Language for our SDK

Natural Language Processing (NLP) is a popular type of artificial intelligence that aides computers to understand human language.

Remarkable advances in cloud computing and machine learning algorithms have unleashed a wave of innovation and very low cost applications for NLP. What was once thought of as a challenging problem in computer science is now a shining accomplishment. Many of the most widely spoken languages and dialects in the world can now be processed by computers very accurately and incredibly fast.

Natural language processing helps companies to analyze, understand, and derive meaning from thousands of words in just a few minutes. Social media monitoring is done almost entirely from NLP without any humans involved in data collection and text analysis. Many large companies use NLP to improve customer support with chatbots or by scanning customer service hotline transcripts for deeper insights.

How does this work exactly? It starts with software developers integrating NLP functions into their applications to do speech recognition, translation, entity recognition, syntax analysis, category segmentation, and, my personal favorite, sentiment analysis.

AimMatic chose Cloud Speech and Cloud Natural Language from Google Cloud Platform™ when we built our SDK for Android developers. The SDK makes it easy to collect user feedback spoken into a microphone using natural language processing.

What are the three reasons why we picked Google?

Google Inc. [CC BY 3.0 (http://creativecommons.org/licenses/by/3.0)], via Wikimedia Commons]

1.Google has seven products that each have over one billion users, including Android with two billion users. Why is this important? It means Google has billions of AI trainers, and more training makes the AI smarter and more efficient.

2. Developers in over 130 countries can use their native language with our SDK. Native language transcripts are more accurate. You can prompt your user to confirm their device language is their native language, and if not, allow them to detect or set the language manually (usually in your app settings). Then you just need to make a microphone UI and incentivize users to give you spoken feedback.

The bottom line is accuracy. You don’t want to waste a word. Accurate transcripts result in accurate analytics that lead to accurate insights from your user base.

From Big Data to Winning [CC BY 3.0 (http://creativecommons.org/licenses/by/3.0)]

3. Cloud Natural Language provides several methods to produce analytics from your transcripts. By default, our SDK always includes three methods: analyzeSentiment, analyzeEntitySentiment, and classifyText. AimMatic produces a suite of tactical insights for you from these three methods, including you can get your Net Sentiment Score (NSS) and Entity Sentiment Scores (ESS).