In this post, we'll use a supervised machine learning technique called logistic regression to predict delayed flights. But before we proceed, I like to give condolences to the family of the the victims of the Germanwings tragedy.

When students enrolled in USC's daunting neuroscience graduate program needed help cracking a tough project, they all went to Bosco Tjan. It didn't hurt that his advice often came with a free cappuccino. Mara Mather, a professor of gerontology and psychology at USC, described Tjan as an affable, caring presence on campus. He always found time to aid students and professors despite a breathless schedule. In many ways, she said, Tjan was the center's heartbeat.

The capabilities of artificial intelligence and machine learning are accelerating, and many cybersecurity tasks currently performed by humans will be automated. There will still be plenty of work to go around so job prospects should remain good, especially for those who keep up with technology, broaden their skill sets, and get a better understanding of their company's business needs. Cybersecurity jobs won't go the way of telephone operators. Take, for example, Spain-based antivirus company Panda Security. When the company first started, there were a number of people reverse-engineering malicious code and writing signatures.

The capabilities of artificial intelligence and machine learning are accelerating, and many cybersecurity tasks currently performed by humans will be automated. There will still be plenty of work to go around so job prospects should remain good, especially for those who keep up with technology, broaden their skill sets, and get a better understanding of their company's business needs. Cybersecurity jobs won't go the way of telephone operators. Take, for example, Spain-based antivirus company Panda Security. When the company first started, there were a number of people reverse-engineering malicious code and writing signatures.

Pokeman Go has been a raging success. But its launch was marred by frequent downtimes and dropped connections. In a recent chat at the Google Cloud Platform Next Conference, Niantic CTO Phil Keslin talked about the "hair on fire" experience where the team had to firefight and upgrade key components on the live production system in order to handle the unexpected surge in new users joining in. Among the various upgrades made to the system, Niantic had to replace the network load balancer with a much more sophisticated HTTP/S load balancing system that could handle a larger overall throughput and offer faster connections. Keslin says that this timely upgrade made it possible for his team to launch in Japan without an incident although the number of new user signups at this point was triple what it was during their earlier US launch.

The federal government's highway safety agency agrees with Google: Computers that will control the cars of the future can be considered their driver. The redefinition of "driver" is an important break for Google. Companies like Visteon (NYSE: VC), which makes infotainment systems and connected car solutions, is already working with leading automakers like Ford, Mazda, Renault/Nissan, GM, Jaguar Land Rover, Honda, BMW, Daimler, PSA and Volkswagen. The company is working on making it easier for automakers to update apps in cars but also add a layer of cyber security to the infotainment system. More apps in cars mean more cybersecurity threats.

According to a set of intelligent humans interviewed for this story, artificial intelligence (AI) and machine learning are going to help drive the tech economy in 2017. When CIO.com posted a query on Help a Reporter Out, a site designed to help journalists connect with sources, asking about startup trends to watch in 2017, the overwhelming majority of respondents pointed to AI. This coming year and beyond, AI will help companies "disrupt sectors that haven't been fully disrupted," says Anthony Glomski, principal of AG Asset Advisory, a financial advisory firm. "AI is in its beginning stages with massive potential impact." Here are eight startup categories and trends experts believe will be big in 2017.

In 2011, cybersecurity researcher and entrepreneur Stuart McClure spent his last year working at McAfee, as the company's global chief technology officer, apologizing a lot. McClure said hackers were slipping into McAfee customer networks and each subsequent breach seemed worse than the last. McClure would have to meet with each of the big McAfee corporate customers to explain why the software failed and at the end of each meeting someone would ask McClure the same question: "'What type of security software do you use on your machine to prevent cyber attacks?'" The customers would then wait, pens poised above a piece of paper to jot down the long list of layer after layer of high-end software that the global CTO of a multimillion-dollar security company would surely recommend. "I would tell them I only trust my brain and my hand, because there are no new ways to breach a network," says McClure of his former employer, which is now owned by Intel.

Machine learning is being used by companies such as Netflix, Facebook, and Spotify for automated data analysis. The results are then used to create recommendations based on past consumption habits. This approach is based on algorithms that learn and adapt to the usage data and patterns that emerge, not the hard-coded rules used in traditional analytics. Netflix in particular has successfully leveraged analytics for years, and their strategy serves as an example of how machine learning can help you gain a competitive advantage. Big Data analytics, being based on manual processes to search for patterns in data, has a major flaw--humans.

This year, before the doors of the the Los Angeles Auto Show open to the public, the show held Automobility, an automotive tech showcase. As our cars become more like two-ton devices that we drive, auto shows are having to adjust their focus to include apps, AI, connected cars, and more. Here are a few of the most innovative tech stories from LA. Hyundai offered Blue Link, an app that allowed owners to unlock and start their cars via smartphone, in 2011. The second generation of Blue Link rolled out in 2014, adding smart watches to the app's repertoire. Now the service works with Amazon's Alexa in-home AI device.