16 Machine Learning Terms You Should Know

Advanced analytics is heating up. AI, machine learning, deep learning, and neural networks are just some of the terms we hear and should know more about. While most of us will never become statisticians or unicorn data scientists, it's wise for us to understand some of the basic terms, especially since we'll be hearing a lot more about machine learning in the coming years. Here are a few terms we should all know from some sites that have much more to offer:

Algorithm - a step by step procedure for solving a problem.

Attribute - a characteristic or property of an object.

Classification - to arrange in groups.

Clusters - groups of objects that share a characteristic that is distinct from other groups.

[Get proactive with your AI and Machine Learning strategy. Sign up now for the next edition of the A2 Academy.]

Correlation - the extent to which two numerical variables have a linear relationship.

Deep Learning - An AI function that imitates the workings of the human brain.

Decision Tree - a decision support tool that uses a tree-like graph or model of decisions and their possible consequences, including chance event outcomes, resource costs, and utility.

Natural Language Processing (NLP) - the automatic (or semi-automatic) processing of human language.

Neural Networks - a series of algorithms that attempts to identify underlying relationships in a set of data by using a process that mimics the way the human brain operates.

Normal Distribution - symmetrical distributions that have bell-shaped density curves with a single peak.

Outlier - an observation that lies an abnormal distance from other values in a random sample from a population.

Regression - a statistical process for estimating the relationships among variables.

Statistical Model - a formalization of relationship between variables in the form of mathematical equations.

Supervised Learning - accomplished with training data that includes both the input and the desired results.

Unsupervised Learning - accomplished with training data that does not include the desired results.

What Terms Would You Add?

This list above does not represent all terms that apply to machine learning because I have limited space and it's more fun to continue the discussion. Please feel free to add some of your favorites in the comments section.

Lisa Morgan, Freelance Writer

Lisa Morgan is a freelance writer who covers big data and BI for InformationWeek. She has contributed articles, reports, and other types of content to various publications and sites ranging from SD Times to the Economist Intelligent Unit. Frequent areas of coverage include big data, mobility, enterprise software, the cloud, software development, and emerging cultural issues affecting the C-suite.

How Today's Analytics Change Recruiting

HR analytics aren't mainstream yet, but they're gaining traction because human capital management is a competitive issue. From recruiting to employee performance and overall HR operations, analytics will have an increasing impact on companies and individuals.

How the IoT Will Impact Enterprise Analytics

Organizations are collecting more data from devices, whether in industrial settings, retail operations, or other installations. Here's a closer look at what they are doing with all that data.

Re: Neural networks
  • 5/13/2017 12:08:29 PM

Theoretically of course those self driving machines should be able to determine all the circumstances of driving safety and avoid injuiries and damage, but as the complexities increase as differening weather and road geometries enter the mix, and not to mention complexities involved when all other cars are assuming like programming, it may be awhile before we'll see total accident free driving for all locations and situations, probably much like the aviation industry where much not is automated but still major accidents occur daily.

Re: Important Terms
  • 5/5/2017 3:18:04 PM

@tomsg, lol.  I'm wondering if it's a generational thing, and I plan to ask that very question in my session at InteropITX which talks about automation and "robots."

Re: Important Terms
  • 5/3/2017 1:01:59 PM

Warning Will Robinson ( with flailing arms) is the real definition of a Robot:)

Re: Neural networks
  • 5/2/2017 10:30:42 AM

Self-driving cars will have the processing power to treat every car or pedestrian around them with the special attention we give to speeding or distracted drivers and intoxicated pedestrians.

In addition, Self-driving cars will have sensors that are watching in all directions at all times, so the processor will have more warning of an unpredictable vehicle nearby.

Of course there will be problems as the implementation is worked out, but the final state should be noticeably better than driving today.

Re: Neural networks
  • 5/2/2017 8:40:43 AM

Assessing the big picture is a struggle that is going to take some time to work through with AI.  Sometimes I think we take the human ability to assess and react for granted.  We see someone wobbling at they're walking down the road and we assume they may be intoxicated and may act unexpectedly so we give them a little extra attention.  Or we watch a car weaving through traffic and we watch for brake lights as that car swerves into our lane.  Self driving AIs know the rules of the road and know to react to data cues but does it know when to switch focus/resources when one cue becomes abnormal?

Re: Important Terms
  • 5/1/2017 5:29:57 PM

I agree with all of you re: common definitions.  I'm having some rather amusing discussions with people about the term "robots" at the moment.  Physical form - not by today's standards necessarily but historically yes, indeed.  It would be easier if robot meant with a physical form and "bot" meant human-impersonating software....but no...;)

Re: Important Terms
  • 5/1/2017 2:50:14 PM

I agree. Maybe we need a wiki or a dictionery of universally accepted terms.

Re: Neural networks
  • 4/30/2017 6:47:07 PM

@Joe I assumed that could be the case, but I didn't realize it was already happening. AI is very far away from handling trickier situations. 

Re: Important Terms
  • 4/30/2017 5:44:19 PM

@rbaz   I couldn't agree more.  Fundamental agreement over the meaning of terms is key to any progress we hope to achieve.

Re: Neural networks
  • 4/30/2017 5:41:56 PM

@Joe   I think you expose the chief barrier to achieving univeral progress when it come to AI. This field will require a level of detail that society has yet to achieve on a consistent basis. One might argue that only NASA and Defense based opperations have come remotely close to achieving this amount of detail on a semi-consistent basis.

This is not to say the private sector cannot achieve this lofty goal, but it will take some time to change the mental approach that most use to address problems and their solution.

Page 1 / 2   >   >>