Organizations gaining new benefits by automating data engineering by Jelani Harper via @infomgmt

A number of advancements have now decreased data preparation time while increasing the time available for exploration and applications.

I think there are several possible tools that enable users to do their own querying and this article talks about the one that the author is most familiar with.  Some organisations use Tableau, Pentaho or Power BI. I'm sure there are others I have not listed.

Security pros complain the cloud obscures compliance issues by Bob Violino via @infomgmt

Half of the organisations surveyed said existing tools aren’t effective in the cloud and an overabundance of tools makes it almost impossible to prioritise IT and security investments.

I can relate to their concerns - I'm not convinced that cloud is mature enough and all the tools are in place to manage all the areas that should be. I think we need more centralisation of security tools and especially those in the cloud area - yes there are some tools that exist as part of cloud management suites but they do not interface or fit in with the other sides of data security and we need something to make sure it can become more central.

Understanding Feature Engineering - 4 part article by Dipanjan Sarkar via @TDataScience

Great 4 part series that you really need to set some time aside so you can sit and read these:

1 - Strategies for working with continuous, numerical data

2 - Strategies for working with discrete, categorical data

3 - Traditional strategies for taming unstructured, textual data

4 - Newer, advanced strategies for taming unstructured, textual data

7 Books to Grasp Mathematical Foundations of Data Science and Machine Learning by Ajit Jaokar via @kdnuggets

It is vital to have a good understanding of the mathematical foundations to be proficient with data science. With that in mind, here are seven books that can help

A great list of books to help with this area.  Certainly I know I need to be better at this.

Understanding fast data and its importance in an IoT-driven world by Kayla Matthews via @infomgmt

Processing high volumes and continuous streams of information in real-time with low to medium latency, it is scalable, has a high uptime and can quickly recover from failure situations.

I think for me you have to have :

- Clean Data - it has to be good data that is not rubbish.
- Data Management - you have to understand exactly what you have and what it means. It has to have consistent definition and there must be some sort of validation to make sure it is correct.
- Process Consistency - your processes have to be consistent too so everyone works off the same thing.

9 key mistakes organizations make when analyzing data by Larry Alton via @infomgmt

The accessibility and ubiquity of information has led to an increased number of amateur mistakes in analysis. Here are some of the most common, and how to overcome them.

I think this list needs to be bookmarked, printed out and more importantly referred to in order to try and check for all of these in order to improve the standard of your analytics.

Overcoming hidden data risks when managing third parties by Baan Alsinawi and Adriaen Morse via @infomgmt

Here are steps that will extend a risk management program to include outside vendors and reduce the likelihood of a breach due to factors outside an organization’s control.

I'd like to think that none of these are new or surprises but recent breaches and legislation (like GDPR) turn a much higher focus on this kind of thing. It's almost a master list for anything that is put out to an external organisation to complete for you.

Understanding the relationship between agile software development and Kanban by Anthony Coggine via @infomgmt

Kanban is not a paradigm in and of itself, but a way to manage projects, tasks and team planning. Projects are done just in time, on a rolling basis.

I'm going to repeat the very last paragraph in this article because it is so important and so clear - "Constant communication, assessment, and reorganization is necessary to complete high quality software projects. Increasing transparency, visibility, and cooperation among team members is the best way to facilitate software development."

I really think we should all print that section out in a large font and put it up where we work in order to make sure we remember it and try to stick to it.

3 steps to help ensure your digital transformation efforts succeed by Justin Rodenbostel via @infomgmt

These projects are risky and 90 percent fail. But the payoff could be market-leading products and services that beat out a close competitor or prevent potential disruptors from stealing market share.

These three are the absolute minimum requirements to try any kind of digital transformation. I think there is a lot of pre work that should be done too.  I think money is crucial to tie down first as far too many projects like this end in a spiral as the cost goes up and up - financial and therefore results must be locked and controlled with a rod of iron to avoid a huge failure.

WEBINAR: How data catalogs can drive real results with business intelligence and analytics - 26 April 2018

Web Seminar  How data catalogs can drive real results with business intelligence and analytics

Apr. 26, 2018 | 2 PM ET/11 AM PT
Hosted by Information Management

Across all industries, and in organizations large and small, everyone wants to do more with data. 

More specifically, everyone wants to use data to get closer to customers, to create new efficiencies, 

to drive smart business decisions, and to boost the bottom line. But that’s a tall order when 

organizations continue to create and collect virtual mountains of information. The challenge is to 

know which data has real value. But too often discussions about data and its value between 

information technology professionals and business stakeholders turns overly technical. 

Business users simply don’t understand what data they possess and where to find it. 

This webinar will look at how data catalogs can ease that process and present data 

assets in ways that everyone in the organization can understand and take advantage of.

George Yuhasz Director, US Business Intelligence & Data Services Keystone Foods (Speaker)	Jay Zaidi Managing Partner AlyData (Speaker)	David Weldon Editor-In-Chief Information Management (Moderator)

Sponsored By:

Register here

If Your Data Is Bad, Your Machine Learning Tools Are Useless by Thomas C. Redman via @HarvardBiz

Poor data quality is enemy number one to the widespread, profitable use of machine learning. While the caustic observation, “garbage-in, garbage-out” has plagued analytics and decision-making for generations, it carries a special warning for machine learning.

Very good advice in this article that I think should be bookmarked or notes should be taken because you really need to be following these steps if you want to be successful with machine learning.

Choose the right AI method for the job by Stephan Jou via @VentureBeat

It’s hard to remember the days when artificial intelligence seemed like an intangible, futuristic concept. Today, AI is everywhere. This has been decades in the making, however, and the past 90 years have seen both renaissances and winters for the field of study.

Some great comments from Stephan.

6 ways to attain top benefits from artificial intelligence & machine learning by Maxim Lukichev via @infomgmt

It can seem overwhelming to choose the right implementation approaches to these hot technologies. Here are six effective ways to attain quantifiable results from AI and ML.

I definitely agree with the 6 ways he has suggested.  I would add a few of my own to his list.

7 Check the data definitions are consistent. For example I have in the past worked at an organisation that called the same data different names in different systems, used different formats for that common data item, or even different values for the same thing in different systems. You need to sort out the MDM of the data first so you always compare apples to apples and pears to pears.

8 Handle the same fields, same data, etc in different physical formats - for example is it text, number, decimal, number. What happens when you convert to a common format - does it change the value or format.

Graph databases and machine learning will revolutionise MDM strategies by Aaron Zornes via @infomgmt

These technologies will become widely adopted in 2018 and 2019, and will augment master data management and data governance to provide increased agility and scalability.

Anything that enhances and improves the understanding of data gets my vote. It is crucial to the production of correct results in any system or reporting that the data and relationships between that data are understood so that any results can be trusted 100%

WEBINAR: Build for the Future of AI and Machine Learning - 19 April 2018

We are at the beginning of a major paradigm shift – from simply acquiring massive amounts of data to harnessing that data with new technologies like artificial intelligence and machine learning at scale. There is so much potential and opportunity for businesses in this new analytic era. The question is: how can businesses successfully adopt and integrate AI and machine learning at scale into their analytic initiatives?   Join us for an interactive virtual event to hear from a panel of analytic experts as they dispel the myths and dive into the nitty-gritty of how AI and machine learning will impact analytic teams.   In this collaborative, peer-to-peer forum we’ll discuss:   How to balance traditional needs with embracing new analytic opportunities   How to build a team ready to take on AI and machine learning   Why building a strong analytic foundation is a critical step in building for a future filled with AI and machine learning   What tools, methodologies, or de-silo-fication will be instrumental in your plans to revolutionize your enterprise data structure   Go from theory to reality with us and start driving analytics success at scale.       April 19, 2018     10:30am PT | 1:30pm ET   Register Now   SPEAKERS   Tom Davenport Co-Founder, Advisor IIA   Dr. Harsh W. Sharma Data Business Group Accenture   Heather Harris Solutions Architect & Data Scientist Alaska Airlines   Olivia Duane Adams CCO Alteryx   Ashley Kramer VP, Product Management Alteryx

Data scientists that produce data-driven products rule the market by Adam Keene via @infomgmt

These professionals are core to the success of software companies, and this role can quickly lead to leadership opportunities and top salary potential.

This gives you some ideas on what is possible in the Data Scientist job area. I would sound a huge caution on some of this - there are many people doing these kinds of role that do not and are not likely to command the salaries suggested in this article.  The best salaries go to those that are in the rock star level of data scientist.

Tired of Social Network Platforms that are affected by bots and trolls? There is an alternative

Tired of Social Network Platforms that are affected by bots and trolls?  There is an alternative

We've all seen the news headlines and reports about mischievous or hostile nations using trolls, bots or data analytics to influence opinions or even elections in other countries and despaired. I'm sure like myself many of you will have thought about recent elections and wondered if the result was correct or even how we got there. Examples are the US Presidential election in 2017 or the UK In or Out of the European Union vote in 2017. No matter the colour of your political beliefs do we really believe all elections have been correct even if we agreed with the result?

But there is an alternative. Hacktivist The Jester has set up a completely new Social Network Platform that blocks contact from any IP space originating in Russia, China, North Korea, Iran, Pakistan or Syria, along with a list of over 100,000 VPN and proxy services. Those abilities combined ensure that the platform is safe and free from any other those negative influences. It is also an environment where discussion is encouraged and all points of view are respected.  New users are welcomed and given suggested hashtags to use to find help (although there will always be someone around to help if you are lost). It is such a safe haven that many users close their other accounts and focus on this new one - even committing to help towards the cost via Patreon or Bitcoin.

So if you want to try something new that encourages respect for the individual and different opinions come and join the discussion on Counter Social here.

Find out more:

Counter Social page

The Jester Wikipedia page

JΞSŦΞR ✪ ΔCŦUΔL³³º¹  on Twitter

WEBINAR: Minimizing Model Risk with Automated Data Preparation & Machine Learning - 19 April 2018

Webinar - Thursday, April 19, 2018 2:00 pm ET/ 11:00 am PT - 60 minutes with Q&A   REGISTER     In today's business landscape, predictive analytics are a necessity to remain competitive, but working with data and developing accurate predictive models is challenging. The quality of predictive output relies on the quality of input. That's why proper data preparation is such a critical success factor for achieving optimal machine learning results. However, getting the data prepared for analysis is a time-consuming process. In addition, models are inherently complex - and if developed poorly can do more harm than good. Register for this webinar to learn how to use Automated Data Preparation & Machine Learning to gain a competitive advantage, while quickly aligning your business operations to regulatory requirements. We discuss current trends and expectations for model risk management regulatory compliance, how to reduce the time it takes to prepare data, and how industry-leading organizations are leveraging Machine Learning to provide a much stronger framework for model development and validation than traditional manual efforts. You'll discover: How Self-Service Data Preparation reduces the work required to get data ready for predictive modeling Efficient methods to organize complex data and marry multiple datasets for predictive modeling How Automated Machine Learning reduces model risk, while ensuring the implementation of cutting edge machine learning models How Automated Machine Learning enhances compliance to model risk management regulation     REGISTER     Speakers       Seph Mard Head of Model Risk Management  DataRobot     Christopher Moore Lead Solution Engineer & Data Wrangler  Trifacta

DataRobot, Inc, One International Place, 5th Floor, Boston, MA 02110 © 2018 Copyright   Contact   Privacy Policy

To protect artificial intelligence from attacks, show it fake data by Jackie Snow via @techreview

Google Brain’s Ian Goodfellow explains how AI systems defend themselves, onstage at EmTech Digital.

A good point that deserves serious thought and consideration.

Who Needs A Data Model Anyway? by @BarryDevlin via @TDWI

Will AI eliminate the need for data models?

I come from a data modelling background so I'm biased but I still think there is a home for a data model. An Enterprise Data Model would be very useful as a tool to help non-technical people understand the business and how the data within it relates.

The rise of blockchain and blockchain-as-a-service by Antonis Papatsaras via @InfoMgmt

The technology creates the perfect conditions for organisations to provide vendors, customers and employees with visibility into their operations.

The potential of this technology is huge but we also need to be really careful not to get blinded by the hype surrounding this technology. You need to both understand it and test it properly so you have clear benefits before implementing.

How artificial intelligence and machine learning can revolutionise ecommerce by Bud Goswami via @InfoMgmt

With the latest advancements in AI, retailers are beginning to really hone in on creating a custom, brand-focused experience for each visitor.

I have to agree with Bud - I think both AI and ML are going to make major differences for all retailers and especially those with large eCommerce bases.

Thanks to Facebook, expect GDPR to spread beyond the EU by Lisa Loftis via @InfoMgmt

The strong and immediate reaction to this data misuse incident should serve as a warning shot for all companies collecting and using consumer personal data.

I agree with this article - the GDPR legislation doesn't seem that difficult now if you want to preserve your reputation and survive going forward.  This Facebook disaster should serve as a lesson to other companies as to what could happen to them if they are not careful.

SLIDESHOW: 5 technologies that will reshape business and society by David Weldon via @InfoMgmt

Within the next five years, blockchain, AI, lattice cryptography and quantum computing will revolutionise software development and its impact, says IBM Research’s global labs.

There are some really exciting developments in this slideshow - I'm particularly looking forward to lattice cryptography.

Learning AI if You Suck at Math by @Dan_Jeffries1 via @hackernoon

"Maybe you'd love to dig deeper and get an image recognition program running in TensorFlow or Theano? Perhaps you're a kick-ass developer or systems architect and you know computers incredibly well but there's just one little problem: You suck at math."

Good suggestions. There are also maths courses on Coursera and other MOOCs.  Of course many tools have functions that you can use to help you get over the maths problem too.

Why AI Cannot Survive Without Big Data by Philip Piletic via @SmartDataCo

Data scientists are struggling to create structure out of the jumble of big data out there – structure that is essential for AI to function properly.

Yes AI needs lots of data, but it also needs to have some kind of known structure to that data as well as an understanding of the meaning of that data.

8 Common Pitfalls That Can Ruin Your Prediction by Norbert Obsuszt by @kdnuggets

A good prediction can help your work and make it easier. But how can you be sure that your prediction is good? Here are some common pitfalls that you should avoid.

This is good and deserves a bookmark so you can refer back to it.

Will GDPR Make Machine Learning Illegal? by Gregory Piatetsky via @kdnuggets

Does GDPR require Machine Learning algorithms to explain their output? Probably not, but experts disagree and there is enough ambiguity to keep lawyers busy.

It sounds like it could be a job creation scheme for lawyers and legislators in the future. I think it might also mean that organisations need to change their privacy policies in order to cover the use of data in ML and get customers to agree to it in order to protect the use going forward.