Senthilkumar G

Q 271. What is a residual in Regression? Why is it important to analyze the residuals before assessing the goodness of a Regression Model? What does it mean if Residuals are non normal or non random?
 
Residual:
Cutting-edge statistics and optimization, Residuals and statistical errors are closely related and easily disordered measures of the deviation of an observed value of an element of a statistical from its “Theoretical Value”.
 
Residual = Observed value - Predicted value
 
The Error of an observed value is the deviation of the observed value from the true value of quantity of interest (for example: a population mean) and the residual of an observed value is the variance between the observed value and the estimated value of the quantity of interest (for example:  a sample mean). The division is most important in regression analysis, where the concepts are sometimes called the regression statistical errors and regression residuals and where they lead to the concept of studentized residuals.
 
Error Vs Residual:
·         The difference between the height of each person in the sample and the unobservable population mean is a statistical error, whereas
·         The difference between the height of each person in the sample and the observable sample mean is a residual.
 
Residual in Regression:
Since a linear regression model is not always appropriate for the data, you should assess the appropriateness of the respective model by defining residuals and examining residual plots.
Residual (e) is the difference between the observed value of the dependent variable (y) and the predicted value (ŷ) and each data point has one residual.
Residual = Observed value - Predicted value
e = y - ŷ
Together the sum and the mean of the residuals are equal to zero ( Σ e = 0 and e = 0).
 
Notation:
e = Residual
y = Observed Value
y’ = Predicted Value
 
Properties:
Σ e = 0
Mean of the residuals e = 0
 
Important to analyze the Residual Plots:
A residual plot is a chart that shows the residuals on the vertical axis and the independent variable on the horizontal axis. The facts in a residual plot are randomly dispersed around the horizontal axis, then linear regression model is mostly appropriate for the particular data set; otherwise, a nonlinear model is more appropriate.
 
The below table shows the inputs and outputs from a simple linear regression analysis.
 

The below chart displays the residual (e) and independent variable (X) as a residual plot.
 

 
 
The above residual plot shows a fairly random pattern, the first residual is positive and then the next two residuals are negative, the fourth one is positive residual, and the last residual is negative. This random pattern is clearly indicating that a linear model provides a moderate fit to the data.
 
Below, the residual plots show three typical patterns for the reference. The following first plot shows a random pattern, indicating a good fit for a linear model.
 
1)     Random Pattern:
 

 
2)     Non-Random: U – Shaped:
 

 
3)     Non-Random: Inverted U
 

 
 
The above last two patterns are non-random (U-shaped and inverted U), suggesting a better fit for a nonlinear model.
 
Residuals are non normal or non random:
Non-normality or non-random of the residual plot is an indication of an inadequate model. It means that the errors the model makes are not consistent cross-ways variables and observations (ie. the errors are not random).
 
Transformations of Variables:
Once a residual plot data set to be nonlinear, it is commonly possible to "transform" the raw data to make it more linear and it will allow us to use linear regression techniques more effectively with nonlinear data.
 
What is a Transformation to Achieve Linearity?
Converting a variable involves using a mathematical operation to change its measurement scale. Generally, there are two kinds of transformations.
 
i)                   Linear transformation. A linear transformation preserves linear relationships between variables. Therefore, the correlation x and y would be unchanged after a linear transformation.
ii)                 Nonlinear Transformation:  Nonlinear transformation changes (increases or decreases) linear relationships between variables and, therefore, changes the correlation between variables.
 
By using Regression, a transformation to achieve linearity is a special kind of nonlinear transformation. The respective nonlinear transformation that increases the linear relationship between two variables.
 
Methods of Transforming Variables to Achieve Linearity:
There are numerous ways to transform variables to achieve linearity for regression analysis. Some common methods are summarized below.
 

Perform a Transformation to Achieve Linearity:
Changing a data set to enhance linearity is a multi-step, trial-and-error process method.
The following steps to be performed for Transforming a data set to enhance Linearity:
i)                   Conduct a standard regression analysis on the raw data.
ii)                  Construct a residual plot.
                      a.     The plot pattern is random, then do not transform data.
                      b.     The plot pattern is not random, then continue.
iii)               Compute the coefficient of determination (R2).
iv)               Choose a transformation method (see above table).
v)                Transform the independent variable, dependent variable, or both.
vi)               Conduct a regression analysis, using the transformed variables.
vii)              Compute the coefficient of determination (R2), based on the transformed variables.
                    a. The transformed R2 is greater than the raw-score R2, then the transformation was successful.
                    b. If not, attempt a different transformation method.
 
The greatest transformation method (exponential model, quadratic model, reciprocal model and etc.) will depend on nature of the original data. Healthier way to determine which method is best is to try each and compare the result (residual plots, correlation coefficients). The finest method will yield the highest coefficient of determination (R2).
 
Reference:
https://en.wikipedia.org/wiki/Errors_and_residuals
https://stattrek.com/regression/residual-analysis.aspx?tutorial=AP
 
 
Thanks and Regards,
Senthilkumar Ganesan,
Email: [email protected]
Mobile: +91-7598124052.
 

Q 261. While Industry 4.0 focuses on usage of new age technology in manufacturing, Service 4.0 focuses on doing the same to provide unparalleled customer experience in the service sector. What are the salient features of Service 4.0?
 
Industry 4.0:
Industry 4.0 is the technology science driven and subset of fourth industry revolution. It encompasses areas which are not normally classified as an industry such as smart industry for instance.
Industry 4.0 is the trend towards automation, data exchange in manufacturing processes which include cyber physical systems (CPS), the Internet of Things (IoT) and Industrial Internet of Things (IIoT), cloud computing, cognitive computing and Artificial Intelligence (AI).
 
Industry 4.0 Revolutions Timeline:

 
 
Industry 4.0 Drivers:
Business and Data analytics are their core capabilities, Industry 4.0 is mainly driven by:
i)  Digitization of products and service offered by respective product owned companies.
ii) Digitization and integration of vertical and horizontal value chains.
iii)Digitization business and business process models and customer access
 
Industry 4.0 Impact and Evolve the Process in Manufacturing:
Scenario: To understand the impact of Industry 4.0 solutions, True transformation happens when all unique challenges and targeting pain points from Manufacturing Personnel in IIoT Ecosystem. The same kind of scenarios are applicable in IoT and Cognitive Computing as well.
a.                 Maintenance Engineer/Maintenance Manager
b.                 Manufacturing Operator/Process Operator
c.                  Plant Engineer/Plant Manager
d.                 Technician, Diagnostics, Repair Installations
 
The four- Step road map for Industry 4.0:
 

 
Industry 4.0 Challenges:
The following challenges are identified in implementation of Industry 4.0.
Economic Challenge:
I)         High Economic costs
II)       Business model adaptation
III)     Unclear economic benefits/excessive investment
 
Social Challenge:
i)         Privacy concerns
ii)       Surveillance and distrust
iii)      Threat of redundancy of corporate IT department
 
Political Challenge:
i)         Lack of regulation, standards and forms of certifications
ii)       Unclear legal/regulatory issues and data security
 
Organizational Challenge:
i)         IT security issues.
ii)       Reliability and stability.
iii)      Integrity.
iv)      Lack of adequate skill-sets to expedite the transition towards the fourth industrial revolution.
 
Service 4.0:
Service 4.0 is a collective term for technologies and concepts of service and support function organizations which based on new disruptive technologies concepts as mentioned below.
i)   Big Data and Mobility
ii)  Internet of Things (IoT)
iii) Internet Services
 
Service 4.0 is an equal capable and similar concept of Industry 4.0 which delivering excellent transformation and its services along with digital capability to various customer in this decade.
Digital transformation with the above disruptive technologies is quite challenge now-a-days to evolve but they will gain sustainable competitive advantage after Service 4.0 transformation that helps companies meet their customer needs.
Successful implementation of Service 4.0 provides the basis for a step change in performance including customer satisfaction beyond traditional lean improvement and reduction in overall costs.
 
Business Strategies:
The following important business strategies are giving best outcome to Industry 4.0 & Service 4.0.  
i)  Strong Customization of Products
ii) Autonomous – Self Optimization, Tuning, Configuration
iii)Self-Diagnosis, Cognition and Intelligent Support
 
Service 4.0 Key Benefits:
The following key benefits are identified for Service 4.0 Transformation:
i)   Greater Flexibility and Adaptability
ii)  Faster, Speed
iii) High Productivity
iv) Better Quality
v)  Performance Improvement
vi) Reduction in Costs.
 
Service 4.0 Offering:
The following important services offered by service providers based on customer needs.
1.     Pro Active.
2.     Integrated, Bundled Package
3.     Customized, Human Centered
4.     Data Driven
 
Assessment or Evolve before Service 4.0 Transformation:
A transformation must address the following four major elements.
1.     Service Providers Operating Model
2.     The Organization
3.     People
4.     IT
 
For Successful Service 4.0 transformation, companies must be in a position to evolve or assess the following:
i)                   Employees/contractors/vendors
ii)                 Hiring New Talents with capabilities of Digital Technologies such Big Data, Cloud and Cyber security and Safety skills.
iii)               Pilots
iv)               Incremental or Agile development approach
v)                 Customer focus or Enabling Process
vi)               Service Transformation initiative
 
Salient futures of Service 4.0:
The following technologies enable the digital transformation and it promote the greater efficiency throughout the value chain.
i)                   Big Data and Analytics.
ii)                 RPA – Robotic Process Automation.
iii)               Bionic Computing. 
iv)               Internet of Things.
v)                 Cloud Computing.
vi)               Cognitive Computing.
vii)             Smart Devices. 
viii)           Virtualization.
 
Industry 4.0 and Service 4.0 are similar concepts and applied to value chain. It promotes greater performance, cost savings and high quality based on customer/consumer needs.  
 
References:
https://en.wikipedia.org/wiki/Industry_4.0
https://en.wikipedia.org/wiki/Service_4.0
https://www.bcg.com/en-in/capabilities/operations/service-4-0-transforming-customer-interactions.aspx
https://www.ibm.com/industries/industrial/resources/smart-manufacturing-optimization/
 
Thanks and Regards,
Senthilkumar Ganesan,
Email: [email protected]
Mobile: +91-7598124052.

Q 258. It is said that the results of the U.S. presidential election in 2016 were influenced by the "filter bubble" phenomenon on user exposure on social media. What is a filter bubble? What are its potential applications in Business World? 
 
What is a filter bubble?
A filter bubble is the result of algorithm and exploration science driven, what information would like to see based on the user’s location, past search behavior and search history of results. This means that results are based only on keywords.
Social sites in current Business World is easily sorted out the user search history based on the algorithm in built and user may experience the filter bubble every time scroll through your favorite news feed on Facebook or search on Google. These social sites could easily determine users most likely to engaging with based on the contents, search history, news feed rather than complete source of information.
Filter bubbles mainly refers to the state of “Intellectual Isolation” that can result from personalized searches on social search engines. With this, users become easily separated from most information that presents on opposing viewpoint. These bubbles are associated with search engines and social media sites.
 
Top Potential Applications in Current Business World:
There are more than 50 top potential applications are currently available in Business World, Users are connecting the world together at same time and browsing that information what they need. 
1.     Google.
2.     Facebook
3.     Amazon
4.     Twitter
5.     MSN
6.     Yahoo
7.     CNN
8.     Netflix and etc.
 
Filter Bubble: The following simple diagram explains about filter bubble and how user searching the information in top potential application in current business world.
In Social sites, if we consider all of these filters and algorithm together and we will understand easily what is filter bubble.
1.     The following diagram which explains about User filter bubble is their own personal and unique universe of information that user lives on online.
 

 
 
2.     In your filter bubble, depends on who you are and what you do.
 

 
3.     Social sites could easily determine which users are most likely to engage with based on the contents, search history, news feed information.
 

 
Beware Online “Filter Bubbles”:
Based on the search engine results (For Ex: Google), we may get a very different search results at the same time while searching on Google search engine. Even when were logged out, there are 57 signals which produced by google to identify mainly following information.
1.     Where user is sitting.
2.     Computer, Laptop or Mobile and Tablets being used by the respective users.
3.     What kind of browser user is using?
Personalized filters have been set for the respective users on social search engines at what you click on first and it can throw off that balance. Instead of a balance information diet, user can end up with surrounded information junk food.
According to Wall Street Journal Analysis, most of the top internet social sites installed on average of 64 data-laden cookies and user personal tracking beacons each. So, before accepting cookies, user need to thoroughly understand the privacy settings and security policy (Data sharing, information security classification and so on) for the respective social sites.
Internet users also should clean up the internet search history, cookies, temp files storage in their respective system on frequent basis.
 
Recommended settings in Individual and Organizational PC’s to avoid Filter Bubbles:
i)                  Periodic review (Individual) & Information Security Audit (Organizational)
ii)                 User Security settings and Clean up.
iii)                Delete Cookies or Block Cookies.
iv)                Delete Search and Watch History, Cache, Staging and Temp files.
v)                 Stay logged out always.
vi)                Encryption mode and Password settings.
vii)               Always focus on what user need rather than entertainment always.
viii)              Limiting the Personalized and Tailored aspects of the Web.
ix)                Custom and Optimized Settings.
x)                 Use the search Engine: SearchEncrypt.com
 
 
Filter Bubble Effect: SearchEncrypt.com
 

 
The user needs to use Private search encrypt engines are a wonderful way to avoid filter bubble. SearchEncrypt.com doesn’t track user search history and the search results based on keywords. Search Encrypt results are not influenced by user political views or history of past internet behavior.
We need the new gatekeepers to encode that kind of public life and sense of civic responsibility/accountability into the code that they’re writing in social sites in current business world so that we connect this world together with safe and secure manner.
 
Reference:
 
https://en.wikipedia.org/wiki/Filter_bubble
https://choosetoencrypt.com/search-engines/filter-bubbles-searchencrypt-com-avoids/
 
Thanks and Regards,
Senthilkumar Ganesan.
Email: [email protected]
Mobile: +91-7598124052.
 
 
Filter Bubble-1.0.docx