Follow :
 
 
 
 
 
 
 

Recommender System – An IntroductionAll slides in PPT(2003)

Recommender Systems –

“…We compare and evaluate available algorithms and examine their roles in the future developments. In addition to algorithms, physical aspects are described to illustrate macroscopic behavior of recommender systems.

Potential impacts and future directions are discussed. We emphasize that recommendation has a great scientific depth and combines diverse research fields which makes it of interests

for physicists as well as interdisciplinary researchers.



Recommender systems with emotional attitudes and preferences

Recommender Systems: Content-based, Knowledge-based, Hybrid

Tutorial: Recommender Systems – A Practitioner’s note book – Practitioner’s gold

http://www.slideshare.net/zh3f/recommender-system-aintroduction

recommenderlab

: A Framework for Developing and

Testing Recommendation Algorithms – You can use the author’s list along with your own development – A meta application of recommender systems

-         A Practitioner’s Gold to apply using R

How to build a recommendation engine

A recommendation engine for product recommendation

A richer, higher intelligence based presentation

http://www.slideshare.net/NYCPredictiveAnalytics/building-a-recommendation-engine-an-example-of-a-product-recommendation-engine?related=1

http://www.slideshare.net/NYCPredictiveAnalytics/recommendation-engine-demystified-4772392

http://bigdata-doctor.com/recommender-systems-101-practical-example-in-r/

A real time recommendation engine in big data based spark system

 
 
For a detailed executive summary see http://www.mckinsey.com/~/media/McKinsey/Business%20Functions/Business%20Technology/Our%20Insights/Disruptive%20technologies/MGI_Disruptive_technologies_Executive_summary_May2013.ashx 

A short meta summary is provided here:

Changes in technologies is always happening due to competitiveness.  There is always hope and excitement when a new technology develops and to top it, the business community looks for opportunities to re-adjust their strategies, and reposition their competitiveness in light of the new developments in technologies.  There are certain technologies that exhibit early explosive successes indicating potential to change the business structures of the world, enhance the quality of lifestyle of people in 10x type form factor, and bring new significant opportunities in untapped areas, and expands the level playing field for all the segments of the world population.  These are considered as disruptive technologies, and in the words of Pete Schumpeter, these lead organizations to “creative destruction” resulting in new industries, new products/services, new production methods, new applications, new lifestyles, bulldozing incumbent businesses with newer ones, resulting in business processes, new business measures, new data, new type of figuring out insights using the new types of data.  We have seen early discussions of such topics as big data in the last 6 years.  Also, the economic impact of these technologies has the range of most possible certainty of $16.7T to possible maximum value of $33T, over shadowing all the previous impacts of all the developments the history of the world has seen.

In this document McKinsey brings out twelve such disruptive technologies, of which the central technology that in fact builds/supports/empowers all the other 11 technologies, is the “Automation of Knowledge Management”, the one that has the maximum of the most certain value, with an economic value of $5.2T, among a total of certainty value of $16.7. That is close to one third of the most certainty economic value that is possible due to all the other 11 disruptive technologies.  The exciting thing is these economic value is realizable in the next 10 years.

This course is uniquely positioned to address the entrepreneurial opportunities that are coming together in the space of data, analytics, and technology, as a platform for knowledge management.

Some key high lights of the knowledge management are:

-          This is all made possible because of the enormous computing power achieved in the last 30 years, a 100 fold increase in the sophistication and the variety of applications.  The iphone 4c is one such a model compared to the computer used in the Apollo 11 lunar module! This makes it possible in the creation of all types of new data as streams, a continuous flow of data resulting in lots of unstructured data, besides structured data, for which we are yet discover efficient and systematic ways of collecting, analyzing, and applying the insights.

-          There will be nearly 230 Million knowledge workers, 9% of the world work force who will be engaged by this disruptive industry

-          The economic impact of these 230 million people will cost nearly $9T, which is one fourth of the global employment costs, indicating the importance of HR/people analytics

-          This will change the nature of work, creates opportunities for entrepreneurs, help create new products and services, changes organizational structures, changes comparative advantages for nations, affects employment, poses new regulatory and legal changes.

Let us get clarity, explore possibilities, and venture in unexplored areas of opportunities.

 
 
 
 
Google maps and Google fusion tables resources are phenomenal.  You can create maps with statistical analysis as well as with pin pointed address based identification with images.  Somewhat similar tool is Microsoft PowerBI and PowerMap tools. But it requires the latest Office product. You are authorized to get Office365 free due to your studentship with Northwestern University.  Go to http://www.it.northwestern.edu/software/microsoft/office365.html  and sign in using the right upper corner pull down menu.  Choose Work/Schoo/University option.  It will allow you to download the full suite of Office365 that includes Excel.  PowerBI is an add on. However, this assignment is a showcase using Google maps and Google Fusion tables.

The R language has a long way to catch up with these tools for its ease and the diversity and the pin-pointed levels of detailed works you can accomplish with latest shape files. So, let us go with Google maps.

The assignment is about usage of county level maps, not more detailed than that.

Step1:  Identify counties with unemployment rate more than 10%, based on the latest numbers released by Bureau of Labor Statistics for December 2015. (50 points)

The following map is an example of counties with unemployment rate more than 10%. Some parts of Kentucky, Mississippi, California seem to be worst hit, according to the latest, DEC15, unemployment rates by counties.  The merged table and the associated map that I created, looks like the one in my blog:

https://www.google.com/fusiontables/DataSource?docid=1FUr92khWVUcnGTwRlqHkLgSmaLOPch1d7is0iCDy

The key learning site you need for your assignment is to use: http://www.poynter.org/2011/how-to-map-data-onto-counties-districts-using-shpescape/141788/  However, attending the special Thursday Sync session (18FEB16) would make this an easier piece to follow.

Download the BLS unemployment rates for December 2015 using the link, http://www.bls.gov/lau/laucntycur14.txt. Download the text file, load it as "|" delimited file into Excel and remove all the other months and save only the DEC-15(P) data to create it as a Fusion table in your google account.  

Download the county level shape files from my directory, https://www.google.com/fusiontables/DataSource?docid=1lyYMQJK9nEo8t1oaHZTaZ86_hTUm2POoBQ6pPk28.

Merge both these accounts using the merge details given here, with in your google account. Details of merge processes are given here: https://support.google.com/fusiontables/answer/2571232#play .  This link also has details on how to create the map in minutes.

Step2: Create a choropleth map(the second map in my link) using the "change feature styles"  on the left navigation panel using all the data, with out any filter (50 points)

Submit the these two sharable link as a one page submission, along with the top page with title page.

 
 
If you have come up to this, this is the destination for communication and hence for visualization...

1. Prisoner of perception limited by the biological limitations is the condition of being human; so is our ability to extend that for fuller picture strengthened by intelligence and empathy.
2. Context is everything
3. Reality is the purpose, but we are driven with what we can process limited by the biological realities and the context of the receiving/presentation layer
4. Hence to realize the reality,
insight is the way; that is a natural graduation to being a human, though a preliminary version of such a talent exists in all living organisms. 


See the video in the link below, where they come to the above conclusions, in a selection of elaborate
presentations.

https://lnkd.in/eUDfZgH

 
 
President Obama, our brilliant President hits home runs on all corners.

On its July 11th Special Communication, in the Journal of American Medical Association (JAMA) President Obama published an article about "
United States Health Care ReformProgress to Date and Next Steps" with the author title Barack Obama, JD. with author affiliates details as,  Author Affiliations - President of the United States, Washington, DC

In simple four graphic illustrations he hits home runs on his key purpose and achievements of ACA.

I copy and post here those four graphics, each comes with its conclusion.  Each of these come with proper citation. 

Some important lessons for students.

Simple line graph, bar graph, linear model graph as very powerful and it will solve most of your key communication needs.
 



 
 
Big data revolution has began

https://www.youtube.com/watch?v=ceeiUAmbfZk


What is big data?

https://www.youtube.com/watch?v=7D1CQ_LOizA

How big is big data?

https://www.youtube.com/watch?v=eEpxN0htRKI 

Human Face of Big Data

https://www.youtube.com/watch?v=r6v15Z60eUI


Demystifying Hadoop

https://www.youtube.com/watch?v=xJHv5t8jcM8

Big data storymap

https://www.youtube.com/watch?v=iFyGuvyesw4 

Analytics:  How Big Data Can Solve our Most Complex Problems

https://www.youtube.com/watch?v=rS1TuTZ2Qw8

Introduction to MapReduce – Map Reduce Programming Tutorial

https://www.youtube.com/watch?v=SoSlCBjJiUA 

Hadoop, cluster computing, and full stack computing architecture

https://www.youtube.com/watch?v=kRnh3WpcKXo

11 Case studies in Big Data Applications

https://www.youtube.com/watch?v=t4wtzIuoY0w 

Value of Big Data – an IT insider View

https://www.youtube.com/watch?v=367KWPASWtE