TDWI | Training & Research | Business Intelligence, Analytics, Big Data, Data Warehousing

Best of Upside's Data Digest

Articles examine data lakes and streams, Hadoop technologies, how to hire data scientists, 5 IoT myths, and big data discrimination.

• By Quint Turner
• June 17, 2016

Use Both Data Lakes and Streams for Best Results
(Source: Information Age)

Data professionals disagree about the relative value of data lakes and data streams. However, this article contends that each has its own niche: data streams for real-time data analysis and data lakes for access to a large variety of stored data.

Important Technologies in Hadoop
(Source: Dbta)

Hadoop is more than big data; it's an entire ecosystem of innovative projects. This article covers 7 types of technology that are influential and impactful in the Hadoop space today.

Big Data Still Requires Human Perception
(Source: Datanami)

Machine learning and big data give people the idea that error-free algorithms will create an error-free world. Unfortunately, as this article points out, these machines are far from perfect. This article explains why the best uses of big data require a mix of machine and human insight.

Best Practices for Hiring Data Scientists
(Source: Icrunchdatanews.com)

Finding the perfect data scientist for your team can be difficult. This article provides 9 helpful tips for hiring a data scientist.

Sensors Sometimes Lie: Busting 5 IoT Myths
(Source: Tech Republic)

From sensors that lie to data that obscures the big picture, this article uncovers some difficult truths about getting value out of IoT projects.

Avoid Accidental Big Data Discrimination
(Source: MarketWatch)

As the number of data sources and variables grows, so does the potential for unintended consequences from big data analytics. If a variable in your algorithm correlates with a protected characteristic (such as race, age, or gender), your results could be discriminatory, and you could be legally liable. This article discusses how to avoid disparate impact when using big data.