Mediaboss Marketing

Mastering machine learning is not easy, even if youre a crack programmer. Ive seen many people come from a solid background of writing software in different domains (gaming, web, multimedia, etc.) thinking that addingmachine learningto their roster of skills is another walk in the park. Its not. And every single one of them has been dismayed.

I see two reasons for why the challenges of machine learning are misunderstood. First, as the name suggests, machine learning is software that learns by itself as opposed tobeing instructed on every single rule by a developer. This is an oversimplification that many media outlets with little or no knowledge of the actual challenges of writing machine learning algorithms often use when speaking of the ML trade.

[Read:How the Dutch government uses data to predict the weather and prepare for natural disasters]

The second reason, in my opinion, are the many books and courses that promise to teach you the ins and outs of machine learning in a few hundred pages (and the ads on YouTube that promise to net you a machine learning job if you pass an online course). Now, I dont what to vilify any of those books and courses. Ivereviewed severalof them(and will review some more in the coming weeks), and I think theyre invaluable sources for becoming a good machine learning developer.

But theyre not enough. Machine learning requires both good coding and math skills and a deep understanding of various types of algorithms. If youre doing Python machine learning, you have to have in-depth knowledge of many libraries and also master the many programming and memory-management techniques of the language. And, contrary to what some people say, you cant escape the math.

And all of that cant be summed up in a few hundred pages. Rather than a single volume, the complete guide to machine learning would probably look like Donald Knuths famousThe Art of Computer Programmingseries.

So, what is all this tirade for? In my exploration of data science and machine learning, Im always on the lookout for books that take a deep dive into topics that are skimmed over by the more general, all-encompassing books.

In this post, Ill look atPython for Data AnalysisandPractical Statistics for Data Scientists, two books that will help deepen your command of the coding and math skills required to master Python machine learning and data science.

Python for Data Analysis, 2nd Edition,is written by Wes McKinney, the creator of the pandas, one of key libraries using in Python machine learning. Doing machine learning in Python involves loading and preprocessing data in pandas before feeding them to your models.

InPython for Data Analysis, McKinney takes you through the entire functionality of pandas and manages to do so without making it read like a reference manual. There are lots of interesting examples that build on top of each other and help you understand how the different functions of pandas tie in with each other. Youll go in-depth on things such as cleaning, joining, and visualizing data sets, topics that are usually only discussed briefly in most machine learning books.Most books and courses on machine learning provide an introduction to the main pandas components such as DataFrames and Series and some of the key functions such as loading data from CSV files and cleaning rows with missing data. But the power of pandas is much broader and deeper than what you see in a chapters worth of code samples in most books.

Youll also get to explore some very important challenges, such as memory management and code optimization, which can become a big deal when youre handling very large data sets in machine learning (which you often do).

What I also like about the book is the finesse that has gone into choosing subjects to fit in the 500 pages. While most of the book is about pandas, McKinney has taken great care to complement it with material about other important Python libraries and topics. Youll get a good overview of array-oriented programming with numpy, another important Python library often used in machine learning in concert with pandas, and some important techniques in using Jupyter Notebooks, the tool of choice for many data scientists.

All this said, dont expectPython for Data Analysisto be a very fun book. It can get boring because it just discusses working with data (which happens to be the most boring part of machine learning). There wont be any end-to-end examples where youll get to see the result of training and using a machine learning algorithm or integrating your models in real applications.

My recommendation:You should probably pick upPython for Data Analysisafter going through one of theintroductory or advanced books on data scienceor machine learning. Having that introductory background on working with Python machine learning libraries will help you better grasp the techniques introduced in the book.

WhilePython for Data Analysisimproves your data-processing and -manipulation coding skills, the second book well look at,Practical Statistics for Data Scientists, 2nd Edition,will be the perfect resource to deepen your understanding of the core mathematical logic behind many key algorithms and concepts that you often deal with when doing data science and machine learning.

But again, the key here is specialization.The book starts with simple concepts such as different types of data, means and medians, standard deviations, and percentiles. Then it gradually takes you through more advanced concepts such as different types of distributions, sampling strategies, and significance testing. These are all concepts you have probably learned in math class or read about in data science and machine learning books.

On the one hand, the depth thatPractical Statistics for Data Scientistsbrings to each of these topics is greater than youll find in machine learning books. On the other hand, every topic is introduced along with coding examples in Python and R, which makes it more suitable than classic statistics textbooks on statistics. Moreover, the authors have done a great job of disambiguating the way different terms are used in data science and other fields. Each topic is accompanied by a box that provides all the different synonyms for popular terms.

As you go deeper into the book, youll dive into the mathematics of machine learning algorithms such as linear and logistic regression, K-nearest neighbors, trees and forests, and K-means clustering. In each case, like the rest of the book, theres more focus on whats happening under the algorithms hood rather than using it for applications. But the authors have again made sure the chapters dont read like classic math textbooks and the formulas and equations are accompanied by nice coding examples.

LikePython for Data Analysis,Practical Statistics for Data Scientistscan get a bit boring if you read it end to end. There are no exciting applications or a continuous process where you build your code through the chapters. But on the other hand, the book has been structured in a way that you can read any of the sections independently without the need to go through previous chapters.

My recommendation:ReadPractical Statistics for Data Scientistsafter going through an introductory book on data science and machine learning. I definitely recommend reading the entire book once, though to make it more enjoyable, go topic by topic in-between your exploration of other machine learning courses. Also keep it handy. Youll probably revisit some of the chapters from time to time.

I would definitely countPython for Data AnalysisandPractical Statistics for Data Scientistsas two must-reads for anyone who is on the path of learning data science and machine learning. Although they might not be as exciting as some of the more practical books, youll appreciate the depth they add to your coding and math skills.

This article was originally published by Ben Dickson on TechTalks, a publication that examines trends in technology, how they affect the way we live and do business, and the problems they solve. But we also discuss the evil side of technology, the darker implications of new tech and what we need to look out for. You can read the original article here.

Published July 8, 2020 09:49 UTC

Here is the original post:
These 2 books will strengthen your command of Python machine learning - The Next Web

Machine learning is critical to effective and scalable virtual care; allowing clinicians to simultaneously improve outcomes and reduce the cost of care.

With the proliferation of sensors and wearables in the home setting, a new host of data is now available for clinicians. And yet, no human can feasibly and economically make sense of this deluge of data. Enter machine learning, which according to Nature, is already showing significant promise augmenting clinicians ability to treat Type II Diabetes, analyze skin lesions, and electrocardiograms. According to Accenture, machine learning will save $150 Billion a year in healthcare costs by 2026.

And yet, todays care model, as is perpetuated by telehealth providers, struggles to adapt and learn from each patient interaction as any learned knowledge that can benefit a population, is effectively lost when clinicians press end call after each session.

Machine learning will unlock clinicians' ability to deliver personalized care at population scale.

Effectively bridging a capacity gap, machine learning is critical to understanding the deluge of data coming from multivariate sensors in the patients home.

Enabling clinicians to scale personalized care to thousands of patients, machine learning will not only allow clinicians to practice top of license it will also foster a learning system that gets smarter with each patient interaction.

Underpinning this opportunity must be an emphasis on transparency and accountability into the drivers of recommendations coming from any machine learning system.

More here:
The value of context and meaning: Virtual care's transformative potential for patient insights - ModernHealthcare.com

A recent study found that two-thirds of health systems have experienced a significant decline in income, with 27% losing money in at least one of the three years studied. The drop in earnings from fiscal year (FY) 2015 to FY 2017 represented $6.8 billion, a startling 44% reduction. There was some improvement in FY 2018, with revenue growth exceeding expenses by 0.1% for the first time since 2015 but growth remains lackluster.

In order for hospitals to advance or maintain standards of care, and to grow in ways that best serve patient needs, they have to get smarter about resource utilization. Operating rooms (ORs) should be a top priority. Not only do ORs represent the majority of a hospitals margin, they are also ripe with ways to increase efficiency and improve capacity. (Read: Top 20 AI Use Cases: Artificial Intelligence in Healthcare)

Multiple factors go into OR utilization: how long procedures are expected to take, which provider is using the room, unforeseen complications that occur during surgeries, last-minute additions or removals to the schedule, and much more. Scheduling ORs is not an exact science, but thanks to advances in data analytics, its getting pretty close.

Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.

New machine learning (ML) solutions can take a variety of data points gleaned from every procedure that happens in every OR in a health system. They analyze this information and make predictions that can be used to effect improvements and adjustments for the future whether its a slight shift in booking times, hiring more staff, or establishing new policies for rooms.

These solutions identify portions of time that can be repurposed, collecting underutilized blocks of time without impacting existing case volume. They also erase countless hours of manual research, conducting queries in mere seconds. (Read: How AI in Healthcare is Identifying Risks and Saving Money.)

Additionally, the insights they yield can save health systems considerable expense, while opening up more treatment opportunities, and thus increase revenue all while improving patient access and the overall patient experience.

Getting the right insights to the right people in OR environments can be an enormous challenge, especially when different groups are looking for answers to different questions. One unified data analytics platform can help, enabling each constituency to find exactly what they need to improve decision-making, which ultimately leads to advancing standards of care, through machine learning.

New machine learning solutions can uncover the various drivers behind high-level trends in volume and utilization across an entire group. For example, from one interface, an executive at a large health group could look at whats happening at a specific hospital or region over a designated time range (such as over the last quarter). The executive might want to generate a holistic view of measures such as OR turnover ratio, add-on ratio, on-time starts, and more.

With this information, it becomes much easier to see how the business of the facility is actually performing. Is it growing? How is its utilization trending?

Based on these results, the executive might want to take a more in-depth look, viewing certain categories year-over-year, grouped by month, in order to compare, spot trends, and get an aggregate historical progression.

Additionally, he could check case volume to look for drivers behind volume increases and which service lines (orthopedics, general surgery, or cardiac care, for example) have experienced the most growth.

Robots are valuable assets, so much so that many groups have committees focused on robots to establish how well they are being used. Robotic program leaders can leverage new data analytics solutions to monitor robot usage and ensure that assets are being fully utilized. With the right insights, committees can figure out if they need to buy additional robots or adjust policies around usage.

Reports can also separate out only robot cases or look at which providers do the most with robots. This helps robotic program leaders know how often ORs equipped with robots are being used for robotics cases only versus other cases, advancing standards of care through machine learning. (Read: What Do Patients Want From Healthcare Technology?)

They also can view specific rooms where robots live by day of the week and hour of the day. Maybe on Mondays a robot is not being used to its maximum potential or a particular provider who isnt using the robot consistently blocks the room. Insights such as these help robotic program leaders figure out where rooms are blocked and why so that they can be leveraged to the fullest.

New solutions can be invaluable to nurse managers as well, giving them the mechanism to make data-driven decisions about staffing across teams and service lines. They can use technology to see historical data (such as room occupancy) to help make future assignments.

Nurse managers might want to view aggregate room data for an entire facility, every hour of each day of the week over the last three months. This way they can spot the busiest (or least busy) times and staff accordingly.

Additionally, a nurse manager may hear from someone on staff that she is always working late. The manager can then run a query to validate if a problem exists by checking times in blocks that are running over and whether certain providers tend to take more time than budgeted. In this instance, the nurse manager could adjust staff hours to keep expectations in line with the likely reality.

These are just some examples of how hospitals are advancing standards of care through machine learning and data analytics, particularly in terms of their ORs. But in order for data to make an impact, it needs to be credible, timely and actionable. If it can be harnessed in this way, insights can change hospitals trajectories.

Data analytics and machine learning unlock the capacity of scarce assets by transforming core processes. Health systems are able to reduce operating costs and potentially defer the need for a facility or staff expansion and most importantly, they can reduce wait time for patients and increase access with greater OR availability. By making operations smarter, the profit-to-expense ratio moves closer to ideal, and organizations can grow efficiently into the future.

The rest is here:
Advancing Standards of Care Through Machine Learning - Techopedia

Rescue teams examine the wreckage of the Ukrainian airliner that was shot down shortly after takeoff in the Iranian capital, Tehran, on Jan. 8. Iran says a tragic series of mistakes led to the missile strike. Akbar Tavakoli/IRNA/AFP via Getty Images hide caption

Rescue teams examine the wreckage of the Ukrainian airliner that was shot down shortly after takeoff in the Iranian capital, Tehran, on Jan. 8. Iran says a tragic series of mistakes led to the missile strike.

Human error, a misaligned missile guidance system and a decision to fire without authorization contributed to Iran's downing of a civilian passenger plane in January, according to a new report from Iran's Civil Aviation Organization.

Ukraine International Airlines Flight 752 was shot down shortly after takeoff from the Iranian capital, Tehran, killing all 176 people on board. For several days, no one took responsibility, but signs pointed toward it being struck by a missile. Eventually, Iran admitted it had targeted the plane after mistakenly believing it was a U.S. missile.

"PS 752 was identified by one of the air defense units as a threat and targeted consequently," Iran's accident investigation report says in its introduction, before laying out the pertinent facts. The Kyiv-bound plane received permission from air traffic control to take off from Tehran's Imam Khomeini International Airport. The flight plan was relayed to Iran's air defense coordination center, and the military granted clearance for the plane to depart.

That's when everything went awry. Mistake No. 1: One of Iran's air defense units had been moved but because of human error, its radar system was never realigned. So when the unit spotted an object traveling on an unknown flight path, the unit didn't realize that it was the Boeing 737-800 that had already gotten clearance.

The air defense unit operator notified command about the object and identified it as a threat. But command never responded and never let the unit operator know that it was the Kyiv-bound flight. "Another link in the chain of events was formed at this point," the report says.

By now, the tragedy still could have been averted, the report says, if not for mistake No. 3: "If at this point he had identified the target as a passenger aircraft, the missile would not have been launched."

The errors continued. From air traffic control's vantage point, PS 752 was following its assigned flight path. But having not heard back from command, the air defense unit operator fired upon the target. He wasn't supposed to do that; Iran's military procedure states that individual units aren't authorized to fire without explicitly getting orders to do so. "The fourth link leading to the firing of the missile was now formed."

The first missile likely hit the plane, the report says. But the rogue unit, still tracking the object, fired again. By now, the aircraft had lost radio communication. It turned to the right. A fire broke out inside the plane. Two minutes later, it crashed into a playground in Khalajabad, exploding on impact. "The aircraft then kept hitting the ground and bouncing on a route towards the airport, making the aircraft pieces, victims' properties, objects and body remains disintegrate completely in a vast area near a residential complex, recreational and sports park, gardens and the surrounding agricultural land," the report says.

All passengers and crew members were killed.

"The sequence of events clearly shows the occurrence of a chain of events initiated by a human error," the report says.

Iran's government had previously explained that it was on high alert after it fired missile strikes on Iraqi bases hosting U.S. troops. That attack was a retaliation for a U.S. drone strike that killed the prominent Iranian military commander Maj. Gen. Qassem Soleimani and militia leader Abu Mahdi al-Muhandis in Baghdad on Jan. 3.

Iran has delayed providing international investigators with the flight's black box recorder but has reportedly pledged to provide it to French investigators on July 20.

Excerpt from:
Rogue Air Defense Unit Fired On Passenger Plane Without Authorization, Iran Says - NPR

It has been 18 months since Kanniyakumari-native Maria Gildas left home for work in Iran. He has been stuck there since, waiting for a seat on a return flight.

A total of 65 fishermen, including 40 from Tamil Nadu, 19 from Kerala and one from Puducherry, are holed up in a room in Iran, waiting for help from the Central and State governments.

A ship brought many home... He did not get a seat on it. We dont have money for a ticket. The last time too, when he tried coming home before the COVID-19 outbreak, he did not have money for a ticket and had to work for two months to earn the money. They now have found a sponsor for the tickets, said Jeeva, wife of Mr. Gildas.

Ms. Jeeva has now pinned her hopes on the kindness and compassion of Chief Minister Edappadi K. Palaniswami.

He has been kind enough to bring back many fishermen. He has written to the Union government asking them to expedite the return of those stranded in Iran, she said.

Mr. Gildas said that they were waiting for details about a flight to Delhi on July 16.

We struggled for a seat on that ship. It had a capacity of only 700 persons, so we were left behind. People here are from several districts, including Nagapattinam and Cuddalore, said Mr. Gildas, the captain of a boat, speaking over the phone to The Hindu.

We are not bothered about eating or sleeping. We just want to return home, he said.

Fisheries Minister D. Jayakumar said that steps were being taken to bring back the men.

They will be brought back at the earliest, he said.

You have reached your limit for free articles this month.

To get full access, please subscribe.

Already have an account ? Sign in

Show Less Plan

Find mobile-friendly version of articles from the day's newspaper in one easy-to-read list.

Move smoothly between articles as our pages load instantly.

Enjoy reading as many articles as you wish without any limitations.

A one-stop-shop for seeing the latest updates, and managing your preferences.

A select list of articles that match your interests and tastes.

We brief you on the latest and most important developments, three times a day.

*Our Digital Subscription plans do not currently include the e-paper ,crossword, iPhone, iPad mobile applications and print. Our plans enhance your reading experience.

Read more here:
Fishermen stranded in Iran look to head back home - The Hindu