Data analytics made accessible pdf

Monday, July 22, 2019 admin Comments(0)

Read Data Analytics Made Accessible: edition PDF Ebook by Anil Maheshwari. Lake Union Publishing, ePUB B00K2I2JL8, SCRIBD. 5 days ago Data Analytics Made Accessible - [Free] Data Analytics Made Accessible [PDF] [ EPUB] Big data is a field that treats ways to analyze. Data Analytics Made Accessible: edition. Data Analytics Made Accessible: edition PDF TagsOnline PDF Data Analytics Made Accessible:

Language: English, Spanish, German
Country: Dominican Republic
Genre: Personal Growth
Pages: 109
Published (Last): 30.09.2016
ISBN: 849-2-31593-391-7
ePub File Size: 21.60 MB
PDF File Size: 9.42 MB
Distribution: Free* [*Regsitration Required]
Downloads: 22246
Uploaded by: EVELYNE

There are many good books in the market on Data Analytics. So, why should anyone write another book on this topic? I have been teaching. Data Analytics Made Accessible: edition - Kindle edition by Anil Maheshwari. Download it once and read it on your Kindle device, PC, phones or tablets. Data Analytics - Made Accessible - Ebook download as PDF File .pdf), Text File ( .txt) or read book online. Data Analytics - Made Accessible.

The difference between a star and snowflake is that in the latter. Data lies at the heart of business intelligence. On the other hand. This equation may include linear and nonlinear terms. Amit Wadhwani rated it it was amazing Feb 14, They can thus be better prepared to fight the diseases. These simple insights can help plan marketing promotions and manage inventory of various movies.

Based on the cross-tabulation above. What is the best selling movie by revenue? What is the best quarter by revenue this year? In this case. The insight that there are more sales of a product in a certain quarter helps a manager plan what products to focus on. These simple insights can help plan marketing promotions and manage inventory of various movies.

What is the best selling geography? What is the worst selling geography? Much effort is required to gather data. Data mining should be done to solve high-priority. One should select the right data and ignore the rest. The value of the insight depends upon the problem being solved. It is important that there be a large expected payoff from finding the insight.

It is possible to try multiple decision-tree algorithms on a data set and compare the predictive accuracy of each tree. The goal is to find a best fitting curve through the many data points.

Artificial Neural Networks: Originating in the field of artificial intelligence and machine learning. ANNs are multi-layer non-linear information processing models that learn from past data and predict future values. Some of the patterns may be more meaningful than the others.

There is no one right answer for the number of clusters in the data. Unlike decision The best fitting curve is that which minimizes the error distance between the actual data points and the values predicted by the curve. These systems also require a large amount of past data to adequate train the system. The data set is divided into a certain number of clusters.

Data can be analyzed at multiple levels of granularity and could lead to a large number of interesting combinations of data and interesting patterns.

Analytics made pdf data accessible

This is most commonly used for market segmentation. Cluster analysis: This is an important data mining technique for dividing and conquering large data sets.

These models predict well. Such highly granular data is often used. This is a well-understood technique from the field of statistics. A retail company may use data mining techniques to determine which new product categories to add to which of their stores.

The user needs to make a decision by looking at how well the number of clusters chosen fit the data. Regression models can be projected into the future for prediction and forecasting purposes. Decision Trees: They help classify populations into classes. There are many popular algorithms to make decision trees. They differ in terms of their mechanisms and each technique work well for different situations.

Here are brief descriptions of some of the most important data mining techniques used to generate insights from data. They use graphs. Present the conclusions and not just report the data. Executive dashboards are designed to provide information on select few variables for every executive. Choose wisely from a palette of graphs to suit the data. Here are few considerations when presenting using data: That is a good reason to prioritize and manage with fewer but key variables that relate directly to the Key Result Areas KRAs of a role.

There is a limit to human comprehension and visualization capacity. An analysis of items frequently found together in a market basket can help cross- sell products. Organize the results to make the central point stand out. Ensure that the visuals accurately reflect the numbers.

These dashboards also have a drill-down capability to enable a root-cause analysis of exception situations Figure 1. Association Rule Mining: Also called Market Basket Analysis when used in retail industry. Make the presentation unique. Inappropriate visuals can create misinterpretations and misunderstandings. Data Visualization As data and insights grow in number. The thickness of the bar shows the number of troops at any point of time that is mapped.

Sample Data Visualization The weather temperature at each time is shown in the line graph at the bottom. One color is used for the onward march and another for the retreat. Time is on horizontal axis. Sample Executive Dashboard Data visualization has been an interesting problem across the disciplines.

The geographical coordinates and rivers are mapped in. Many dimensions of data can be effectively displayed on a two-dimensional surface to give a rich and more insightful description of the totality of the story. It covers about six dimensions. Chapter 8 will describe how Cluster Analysis can help with market segmentation.

Chapter 3 will briefly explain what is data warehousing and how does it help with data mining. Chapter 6 will describe statistical regression modeling techniques. Chapter 10 will introduce the concepts and techniques of Text Mining.

The rest of the book can be considered in three sections. Section 1 will cover high level topics. Section 2 is focused on data mining techniques. Chapter 4 will then describe data mining in some detail with an overview of its major tools and techniques. Chapter 13 has been added as a primer on Data Modeling. Chapter 2 will cover the field of business intelligence and its applications across industries and functions.

Chapter 5 will show the power and ease of decision trees. Section 3 will cover more advanced new topics. Chapter 11 will cover provide an overview of the growing field of web mining. Organization of the book This chapter is designed to provide the wholeness of business intelligence and data mining.

Every technique will be shown through solving an example in details. Chapter 12 will provide an overview of the recent field of Big Data. Chapter 7 will provide an overview of artificial neural networks. Review Questions 1: Describe the Business Intelligence and Data Mining cycle.

What are the similarities between diamond mining and data mining? What are the different data mining techniques? Which of these would be relevant in your current work? What is a dashboard? How does it help? Create a visual to show the weather pattern in your city. Describe the data processing chain. Could you show together temperature. Chapter 4 will describe data mining as a whole. Chapter 2 will cover business intelligence concepts. Chapter 3 will describe data warehousing systems.

Chapter 5 will describe data visualization as a whole. Section 1 This section covers three important high-level topics. Businesses use many techniques for understanding their environment and predicting the future for their own benefit and growth. Data-based decisions are more effective than those based on feelings alone. A skilled business person is motivated to use this cache of data to harness nature. It can be mined for value. There is a new sense of importance and urgency around data as it is being viewed as a new natural resource.

BIDM cycle The nature of life and businesses is to grow. Figure 2. Chapter 2: Decisions are made from facts and feelings.

Business Intelligence and Data Mining Made Accessible

Information is the life-blood of business. In a hyperconnected world. Actions based on accurate data. With this kind of a resource classrooms are being flipped … i. Students can access the lessons at any time to learn at their own pace. It shot into prominence when Bill Gates promoted it as a resource that he used to teach his own children. Khan Academy — BI in Education Khan Academy is an innovative non-profit educational organization that is turning the K education system upside down.

Teachers are provided a set of real-time dashboards to give them information from the macro level "How is my class doing on geometry? It provides short YouTube based video lessons on thousands of topics for free. Khan Academy has developed tools to help teachers get a pulse on what's happening in the classroom. How does a dashboard improve the teaching experience?

Design a dashboard for tracking your own career. BI for better decisions The future is inherently uncertain. The goal is to make effective decisions. The speed of action has risen exponentially with the growth of the Internet. Risk is the result of a probabilistic world where there are no certainties and complexities abound. Reliable knowledge about the future can help managers make the right decisions with lower levels of risk. People use crystal balls. Businesses calculate risks and make decisions based on a broad set of facts and insights.

On the other hand. Research has shown that an unfavorable comment about the company and its products on social media should not go unaddressed for long. In a hypercompetitive world. The Internet and mobile technologies allow decisions to be made anytime.

In that case. BI can help automate operations level decision-making and improve efficiency by making millions of microlevel operational decisions in a model-driven way. For example. Strategic decisions are those that impact the direction of the company.

The decision to reach out to a new customer set would be a strategic decision. Effective BI has an evolutionary component. Operational decisions are more routine and tactical decisions.

Current business models can be tested against the new data. BI can also help create new ideas based on new patterns found from data mining. Developing such decision tree models is one of the main applications of data mining techniques. BI can help make both better. Updating an old website with new features will be an operational decision. An unending process of generating fresh new insights in real time can help make better decisions.

Operational decisions can be made more efficient using an analysis of past data. A decision-tree-based model could provide a consistently accurate loan decisions. In strategic decision-making. A classification system can be created and modeled using the data of past instances to develop a good model of the domain.

BI can help with what-if analysis of many possible scenarios. Decision types There are two main kinds of decisions: The consequences of the decision would be apparent some time later. When people and organizations act. This model can help improve operational decisions in the future. The analytical features include basic statistical and financial functions. A spreadsheet tool. At the user end. BI tools include data warehousing.

BI tools can range from very simple tools that could be considered end-user tools. BI Tools BI includes a variety of software tools and techniques to provide the managers with the information and insights needed to run the business. Pivot tables help do sophisticated what-if analysis. This system offers limited automation using macros and other features. Even executives can be their own BI experts. The dashboards are linked to data warehouses at the back end to ensure that the tables and graphs and other elements of the dashboard are updated in real time Figure 2.

Information can be provided about the current state of affairs with the capability to drill down into details. Data can be downloaded and stored in the spreadsheet. A dashboarding system. The back-end data analytical capabilities include many statistical functions. Add-on modules can be installed to enable moderately sophisticated statistical analysis. Open source systems. Sample Executive Dashboard Data mining systems. The problem should be valuable enough that solving it would be worth the time and expense.

It takes a lot of time and energy to gather. A good data mining project begins with an interesting problem to solve.

[Pdf] data analytics made accessible edition read book anil ma…

The problem needs to be looked at from a wider perspective to consider many more angles that may not be immediately obvious. A skilled and experienced BI specialist should be open enough to go outside the box. Selecting the right data mining problem is an important skill. BI Skills As data grows and exceeds our capacity to make sense of it.

The skill level has to be deep enough to engage with the data and make it yield new useful insights. An imaginative solution should be proposed for the problem so that interesting and useful results can emerge.

The data miner needs to persist with the exploration of patterns in the data. Even a customer complaint can be seen as an opportunity to wow the customer.

BI Applications BI tools are required in almost all industries and functions. By segmenting the customers. The following are some areas of applications of BI and data mining. Identify and delight highly-valued customers. Scoring each customer on their likelihood to quit.

Maximize the return on marketing campaigns: The nature of the information and the speed of action may be different across businesses. BI applications can impact many aspects of marketing. A happy customer becomes a repeat customer.

Pdf made accessible data analytics

Loyalty programs can be managed more effectively. They can be proactively contacted. Maximize customer value cross-.

A business should understand the needs and sentiments of the customer. Offering a customer new products and solutions based on those imputed needs can help increase revenue per customer. Customer Relationship Management A business exists to serve a customer.

Businesses need to embed new insights into their operating processes to ensure that their activities continue to evolve with more efficient practices.

Improve customer retention churn analysis: It is more difficult and expensive to win new customers than it is to retain existing customers. Every contact with the customer should be seen as an opportunity to gauge their current needs.

A business can create a listening post to listen to social media chatter about itself. Treatment effectiveness: The prescription of medication and treatment is also a difficult choice out of so many possibilities.

Data Analytics - Made Accessible

This makes diagnosis as much of an art form as it is science. Decision trees can help doctors learn about and prescribe more effective treatments. Wellness management: This includes keeping track of patient health records. Accurately diagnosing cases of cancer or diabetes can be a matter of life and death for the patient.

Manage fraud and abuse: Some medical practitioners have unfortunately Manage brand image. Healthcare and Wellness Health care is one of the biggest sectors in advanced economies. It can then do sentiment analysis of the text to understand the nature of comments. Evidence- based medicine is the newest trend in data-based health care management.

BI applications can help apply the most effective diagnoses and prescriptions for various ailments. These systems take away most of the guess work done by doctors in diagnosing ailments. They can also help manage public health issues. Diagnose disease in patients: Diagnosing the cause of a medical condition is the critical first step in a medical engagement.

There are also interactions in terms of which drugs work well with others and which drugs do not. Fund-raising from Alumni and other donors: Schools can develop predictive models of which alumni are most likely to pledge financial support to the school.

Google has been known to predict the movement of certain diseases by tracking the search terms like flu. Schools can develop models of what kinds of students are attracted to the school. Retail Retail organizations grow by meeting customer needs with quality products. Understanding emerging By using effective forecasting tools and techniques. Public health management: The management of public health is one of the important responsibilities of any government.

The students at risk of not returning can be flagged. This could lead to a reduction in the cost of mailings and other forms of outreach to alumni. Education As higher education becomes more expensive and competitive. There is a strong need for efficiency.

Exception reporting systems can identify such providers and action can be taken against them. They can thus be better prepared to fight the diseases. Course offerings: Schools can use the class enrolment data to develop models of which new courses are likely to be more popular with students.

This can help increase class size. Schools can create a profile for alumni more likely to pledge donations to the school. Student Enrollment Recruitment and Retention: Marketing to new potential students requires schools to develop profiles of the students that are most likely to attend. Promotional discounted product bundles can be created to push a nonselling item along with a set of products that sell well together.

Banking Predicting sales trends dynamically can help retailers move inventory to where it is most in demand. Optimize logistics for seasonal effects: Seasonal products offer tremendously profitable short-term sales opportunities. If it is raining in a certain area. Optimize inventory levels at different locations: Retailers need to manage their inventories carefully. By tracking sales trends. Retailers generate a lot of transaction and logistics data that can be used to diagnose and solve problems.

This knowledge of affinities between products can help retailers co-locate those products. Understanding which products are in season in which market can help retailers dynamically manage prices to ensure their inventory is sold during the season. Carrying too much inventory imposes carrying costs. Minimize losses due to limited shelf life: Perishable goods offer challenges in terms of disposing off the inventory in time. Improve store layout and sales promotions: A market basket analysis can develop predictive models of which products sell together often.

Retail organizations can provide their suppliers with real time information about sales of their items. Predict changes in bond and stock prices: Forecasting the price of stocks and bonds is a favorite pastime of financial experts as well as lay people.

Financial Services Stock brokerages are an intensive user of BI systems. They also want to retain more good customers. Stock transaction data from the past. Exception-seeking models can identify patterns of fraudulent transactions. This can help traders develop long- term trading strategies. Selling more products and services to existing customers is often the easiest way to increase revenue. Banks make loans and offer credit cards to millions of customers.

Using past data and trend analysis. Optimize cash reserves with forecasting. Banks have to maintain certain liquidity to meet the needs of depositors who may like to withdraw money. Fortunes can be made or lost based on access to accurate and timely information. A checking account customer in good standing could be offered home. Detect fraudulent transactions: Billions of financial transactions happen around the world every day.

These can be inserted in business processes to automate the financial loan approval process. Automate the loan application process: Decision models can be generated from past data that predict the likelihood of a loan proving successful.

They are most interested in improving the quality of loans and reducing bad debts. Monetary policy changes such as Federal Reserve interest rate change or geopolitical changes such as war in a part of the world can be factored into the predictive model to help take action with greater confidence and less risk.

Optimize marketing to specific customers: By micro-segmenting potential customers. Decision models using decision trees can be created to assess the impact of events on changes in market volume and prices.

Assess the effect of events on market movements. By using the best available data to model the likelihood or risk of such events happening. Identify and prevent fraudulent claim activities. Patterns can be identified as to where and what kinds of fraud are more likely to occur.

Decision-tree-based models can be used to identify and flag fraudulent Insurers use actuary tables to project life spans and disease tables to project mortality rates. Forecast claim costs for better business planning: When natural disasters. Determine optimal rate plans: Pricing an insurance rate plan requires covering the potential losses and making a profit.

Identify and prevent fraudulent activities in trading: There have unfortunately been many cases of insider trading. Fraud detection models seek out-of-the-ordinary activities. Progressive Insurance is a US-based company that is known to actively use data mining to cherry pick customers and increase its profitability.

Insurance This industry is a prolific user of prediction models in pricing insurance proposals and managing losses from claims against insured assets. Millions of such customer calls happen every month. Churn management: Telecom customers have shown a tendency to switch their providers in search for better deals.

Discover novel patterns to improve product quality: Quality of a product can also be tracked. Data mining can help with root cause analysis that can be used to identify sources of errors and help improve product quality in the future. Telecom BI in telecom can help with the customer side as well as network side of the operations. Key BI applications include churn management. Preventive maintenance can be planned.

Many companies. The telecom companies need to provide a consistent and data- based way to predict the risk of the customer switching. Decision models to forecast machinery failures could be constructed using past data.

The level of risk should to be factored into the kind of deals and discounts that should be given. From machines working right. Telecom companies tend to respond with many incentives and discounts to hold on to customers. Manufacturing Manufacturing operations are complex systems with inter-related sub- systems.

Predicting which machine is likely to shut down is a complex process. In addition to customer data. Fraud Management: There are many kinds of fraud in consumer transactions. Modeling the failure pattern of various components of the network can help with preventive maintenance and capacity planning. Los Angeles Police Department LAPD mined the data from its 13 million crime records over 80 years and developed models of what kind of crime going to happen when and In telecom infrastructure.

Law enforcement: Social behavior is a lot more patterned and predictable than one would imagine. Public Sector Government gathers a large amount of data by virtue of their regulatory function. A decision-tree- or a neural network-based system can be used to guide the customer-service call operator to make the right decisions for the company. That data could be analyzed for developing models of effective functioning. An American telecom company.

Marketing and product creation. Superimposition fraud involves illegitimate activity by a person other than the legitimate account holder. There are innumerable applications that can benefit from mining that data. Network failure management: Failure of telecom networks for technical failures or malicious attacks can have devastating impacts on people.

A couple of sample applications are shown here. Subscription fraud occurs when a customer opens an account with the intention of never paying for the services.

Decision rules can be developed to analyze each CDR in real time to identify chances of fraud and take effective action. Scientific research: His chief aim is to invite the readers to join the field of data science. I did not like how brief most of the chapters turned to out to be.

Though this is meant to be an introductory book, the author is too brief in some of the chapters and thus the concepts a left floating on the readers head. This is more prevalent in chapters towards the end of the book. This is the best introduction to data science book available. The author uses simple language to discuss a fairly complex topic and covers a wide spectrum of data science concepts.

I would highly recommend the book to those very new to the field.

This is not recommended for anyone with a fair amount experience. Mar 19, Wes rated it it was amazing. Good Overview Good high level overview on a large number of topics. Doesn't get too detailed but uses good examples to demonstrate concepts. Good for a basic understanding of the breadth of data analytics. Nov 11, Cyndi rated it it was amazing.

I found this book very informative. It filled in some of the gaps I had in understanding data science. I particularly liked that it explained some of the statistics involved. This was a good read and everything was understandable. I feel ready for my Data class. Good for beginners This is a good book for people who are just learning about data analytics and all its components. Each chapter was one aspect of data analytics and very informative with examples to illustrate the points.

Feb 24, Christian Ruiz rated it it was amazing. Super completo. Dec 31, Bari Dzomba rated it really liked it. On a scale of with 1 being an introductory book this is about a 2. Sep 02, Gonzalo A Gomez A rated it it was amazing. Great and concise book Not just the book was great, the questions, examples and exercises were amazing. I truly recommend this book to anyone is planning to embark in an data analytics endeavor.

Jun 24, Ulugbek Normatov rated it it was amazing. Amazing book The author of the book knows this subject very deeply, easy to understand , well explained. I recommend this book to everyone who wants to understand big data theoretically and practically. Feb 29, Rick Yvanovich rated it really liked it Shelves: This is a sort of everything you want to know about Business Intelligence and Data Mining.

Lots of bullet form, and checklists of knowledge you need to remember. Relatively easy read despite the subject matter. Amit Wadhwani rated it it was amazing Feb 14, John Bethke rated it liked it Oct 17, Josh Hankins rated it liked it Nov 13, Matt Wiediger rated it really liked it Jul 09, Kingsley Mbua rated it it was amazing Apr 09, Andrew Toh rated it really liked it Nov 01, Aldo rated it really liked it Nov 25, Kerri Millett rated it really liked it Oct 02, Kob Croak croak rated it really liked it Mar 16, Alexander rated it really liked it May 11, Taghrid rated it liked it Jun 21, Are you sure you want to Yes No.

Be the first to like this. No Downloads. Views Total views. Actions Shares. Embeds 0 No embeds. No notes for slide. Data Analytics Made Accessible: This book fills the need for a concise and conversational book on the growing field of Data Science. Easy to read and informative, this lucid book covers everything important, with concrete examples, and invites the reader to join this field.

The chapters in the book are organized for a typical one-semester course. The book contains case-lets from real-world stories at the beginning of every chapter. There is also a running case study across the chapters as exercises. This book is designed to provide a student with the intuition behind this evolving area, along with a solid toolset of the major data mining techniques and platforms.