[DZone Research] Popular AI Algorithms and Their Uses

This article is part of the Key Research Findings from the DZone Guide to Artificial Intelligence: Automating Decision-Making.

Introduction

As part of the research for our 2018 Guide to Artificial Intelligence, we surveyed 403 developers, data scientists, and technologists. From their responses, we've created a quick article on the different algorithms available for working with various AI projects, and which one proved more popular.

Feeling the Algorithm

When asked which algorithms they use for machine learning, respondents gave four main answers: linear regression, decision trees, neural networks, and natural language processing (NLP). Linear regression received the most interest in this survey, with 37% of respondents using this algorithm. Decision trees were a close second at 35%, neural networks were third with 31%, and NLP received 29% of respondents' votes.

Though NLP did not rank as high as expected in our algorithms poll, natural language processing and neural networks are two of the more popular algorithms in the fields of AI and machine learning. Thus, we asked respondents how they are using these two important algorithms.

50% of respondents who use neural networks in their projects use them for image processing. The second most popular use for neural nets, with a 45% adoption rate among survey takers, was to make inferences. Some other popular uses of this algorithm were time series forecasting (38%), making non-linear connections (33%), and character recognition (33%).

For the respondents who use NLP, 61% utilize this algorithm in their chatbots and 60% for parsing text into useful data. These were by far the most popular uses for NLP in this year's survey. This third most popular option was for summarizing complicated data (31%) and for use in search engines (28%).

Conclusion

In our last article, the data collected from our survey correlated rather well with the likes and dislikes of the wider development community. What about our data on algorithms? 

In a great post titled, "A Tour of the Top Ten Algorithms for Machine Learning Newbies," the author discusses several of the algorithms mentioned by our community. The number one algorithm on this list, just like ours, was linear regression. 

For use cases, however, our data seems to diverge from the wider developer community.  In Stack Overflow's 2018 Developer Survey, developers rated the automation of tasks as the most exciting use case for AI. 

This article is part of the Key Research Findings from the DZone Guide to Artificial Intelligence: Automating Decision-Making.