L1: Analytics Tools & Techniques (Lecture 1)

optimization

Data analytics: types

How can we achieve the best outcome

predictive modeling

Data analytics: types

What will happen next if?

simulation

Data analytics: types

What could happen happen...?

forecasting

Data analytics: types

What if these trends continue?

forecasting

-using historical data to see patterns and determines the next years/months

-ex: electrical bill of bulsu

alerts

Data analytics: types

What actions are needed?

alerts

-used for fraud detections

-ex: if there are unusual event/transactions in the credit card, the owner gets notified through analyzing historical transactions of the owner

query/drill down

Data analytics: types

What exactly id the problem?

query/drill down

-root causing

-finding roots/ root causes

-used when it greatly affect sales

-also includes diagnostic

ad hoc reporting

Data analytics: types

How many, how often, where?

ad hoc reporting

-requested reports

-whenever needed reports

standard reporting

Data analytics: types

What happened?

standard reporting

-used/usually provided in regular interval

-ex: annual/monthly report

-expecting figures

-expected to be submitted always

descriptive analytics

-query/drill down: What exactly id the problem?

-ad hoc reporting: How many, how often, where?

-standard reporting: What happened?

predictive analytics

-predictive modeling: What will happen next if?

-simulation: What could happen happen...?

-forecasting: What if these trends continue?

-alerts: What actions are needed?

predictive analytics

-decision tree

-neural network

-support vector machine

-linear regression

-bayesian network

prescriptive analytics

-optimization: How can we achieve the best outcome

prescriptive analytics

more on optimization- maximizing resources to its gain & best possible outcomes/options

prescriptive analytics

provides best action to perform

Operation Research (OR)

prescriptive analytics

doing more with the data you have and making better decisions

Optimization

prescriptive analytics

using a set of mathematical techniques to find the best possible solution to a business problem

big data

classify data through 5v's (volume, veracity, value, velocity, variety)

big data

to store and process through using hadoop, cassandra and spark

big data

helps in disaster management through prediction

big data

we are living in a world with vast amount if information

volume

amount of data

velocity

amount of time to transport data from one medium to another medium

veracity

reliability of the provided information

variety

structured, unstructured, semi-structured

value

if this information is used for model to identify possible outcomes

value

after properly analyzing and extract useful information

data rich

-abundance of data that we have

-sensor, QRs

information poor

-lacking of actionable insights

-the gap is: lacking of skills and tools

to overcome being information poor

-collaboration to beat data silos

-training for data interpretation skills

-investment in analytics

jamovi

-for descriptive analytics

-good tool for describing data

OpenRefine

- performing data transformation

-for addressing inconsistencies of data

analytics

ability to use/extract data into useful insights

analytics

process starts with:

1. data

2. insights

3. action

actionable insights

2 keys in analytics

analytics does not stop by simply generating insights but to have insights that will drive _________________

decision making

2 keys in analytics

ensure that the fata/raw data you use is of high quality to generate reliable insights as a foundation in ____________________

why data driven decision making is important?

-continual business growth

-new business opportunities

-knowledge & innovation

-enhanced communication

-unrivaled adaptability

3 major types of analytics

-descriptive analytics

-predictive analytics

-prescriptive analytics