ISM 3541 FSU Exam 1

Analytics

Ties the landscape together and allows a person to decompose the ecosystem into manageable components

Data Modeling

Descriptive Modeling

Leveraging known, historical data to create a story about how an object (e.g. think noun) arrived in its current state

Predictive Modeling

Creating modules on objects based on attributes gathered from data sets to determine the future behavior of an object and being able to rationalize the prediction based on a formula or algorithm that incorporates risk and probability. The type of model for forecasting one path, and one path only based on assumptions. The date matures as the model runs in iterative fashions.

Prescriptive Modeling

Analytical modeling that supports complex forecasting that is designed to provide insights on decision making. The type of model used for forecasting multiple paths and advising problem solving construct using "what if" scenarios and risk analysis.

Descriptive Analysis

is "predictive analytics" on steroids where multiple paths for future behavior are guesstimated based on probability, risk, impact, and priority and is supported by an overall recommendation(s). Type of model for understanding the past

why/what

when

how

The "schedule-scope-resources" defines the _______, the _______, and the _______.

Project Management Body of Knowledge (PMBOK)

A process for defining the "schedule-scope-resources" boundaries that evolve through a concept of progressive elaboration over a period of time defined by a hard start and end date.

Schedule Performance Index (SPI)

The project management analytic that speaks to the overall timeline. Is a ratio of the earned value (EV) to the planned value (PV). If the SPI is less than one, it indicates that the project is potentially behind schedule whereas a Schedule Performance Index of greater than one, indicates the project is running ahead of schedule.

Cost Performance Index (CPI)

It measures the value of work completed compared to the actual cost spent on the project. As per the PMBOK Guide, "The CPI is a measure of the cost efficiency of budgeted resources, expressed as a ratio of earned value to actual cost."

Earned Value (EV)

is the value of work performed expressed in terms of the approved budget assigned to that work for an activity or WBS component.

Planned Value (PV)

Planned Value is the approved value of the work to be completed in a given time. It is the value that should have been earned as per the schedule. As per the PMBOK Guide, "PV is the authorized budget assigned to work to be accomplished for an activity or WBS component."

Progressive Elaboration

a concept acknowledging the learning process humans engage as projects are executed where as time progresses, the person gains greater level of detail. The evolution of context with time is the reason the analytics are gathered to measure the estimated time versus the actual time and as we become more accurate at defining and estimating work then the data will show our effectiveness.

Actual Cost

A project management analytical attribute that provides insights into the actual cost of an activity.

Database Software

Stores data

Data Structures

Important in the analytics world because they ultimately influence our answers, our assumptions, and the overall reliability of the analytics model.

Weather Channel, Bank of America portal, Dunkin Donughnuts app, or reports built on queries (SQL)

Give examples of applications that are designed to extended data into information and sometimes information into knowledge.

Expert System

Information that extends to knowledge or wisdom

Knowledge Management Systems

Data extended to information and information extending to knowledge

Management Information Systems

Data extends to information

Big Data (data science)

A strategy for many businesses and should be considered when developing a company's vision for where they are heading.

App

An executable file that is stored locally on a device and conducts some processing of data locally and then passed back to the server

Web Application

Implies a browser-based form that processes data through external servers and presents the results on the device's screen

The two important factors are (a) the firm's management must think data analytically and (b) the management must create a culture where data science methods, and data scientists, will thrive.

How does a business ensure that it gets the most from the wealth of data?

Knowledge Management Objects

Where the customer (e.g., object that is a noun) is both a consumer of product/service and a supplier of valuable information.

To gather data as a long-term strategic opportunity by leveraging Big Data

In today's business environment, the real opportunity and value is what?

Protect unique intellectual property

Data science intellectual property can include novel techniques for mining the data or for using the results.

Protect unique intangible collateral assets

Your model is not what your data scientists design, it's what your engineers implement; therefore, the actual source of good performance of a successful data science solution is unclear to competitors.

Hire superior data scientists and create a culture

For all types of professionals to thrive

Big Data

A critical capability for those interested in connecting with their customer base and gaining insight into what motivates their behavior, what they think, and what they value. Big Data is conceptually the data store, or the database, that is mostly decentralized and complex in today's world, especially when building the method to combine multiple data sources together.

Supervised Learning Method

In this learning style, the learner has access to data sets (e.g., training data) and creates and runs models off their interpretation of the data. Models are constantly adjusted as more data is included in the model.

Unsupervised Learning Method

This learning style is true research and development, where we have little clues in terms of the data and what the data represents. We are applying tools and techniques designed to take data from the subjective to objective perspective. This is the difference of opinion and fact. This type of style requires unbounded time and resources in order to evolve the model to take advantage of the learning that occurs.

Classification and Logistical Regression

Two types of data mining methods that are both considered supervised learning.

Classification

A supervised method focused on predicting the target variable of a new, unseen instance by determining which segment that the data point resides.

Logistical Regression

Another supervised method has to do with predicting and applying probability to our model. A technique used for data sets that have a binary attribute of interest, which is either one value or the other. It handles the unknown variables.

-A probability typically ranges from zero to one.

-Odds: Applying the notion of the likelihood of an event.

-When applying the odds, data scientists are interested in the interpretation of the distance from the separating boundary by taking the logarithm of the odds called the "log-odds."

Logistical regression terms and guidelines that are important to understand.

-Analytic thinking

-A culture that encourages deep learning through research

Factors that increase the wealth of data _________ and _________ are the two major factors presented that would maximize the value of an organization's data

The dot.com era

What historical circumstances now exist that may not continue indefinitely?

Declining costs of storage and the increased options of storing information in a cloud environment.

What historical circumstances allow me to gain access to or to build a data asset more cheaply than will be possible in the future?

A first mover advantage of being a low-cost reseller of books was unique for the moment and ultimately moved to the low-cost reseller of all "things"

Which circumstances will allow me to build a data science team now that would be very costly (or impossible) to build in the future if the company waits?

Protect Unique Intellectual Property

Data science intellectual property can include novel techniques for mining the data or for using the results

Protect Unique Intangible Collateral Assets

The actual source of good performance of a successful data science solution is unclear to competitors

Hire superior data scientists

Build a culture through talent

Networks

Big Data Concepts, Writing through Big Data occurs through _______.

Furthest

Analyzing Data, When analyzing data and establishing decision boundaries, we can conclude that the data from the boundary is rooted deeply in their classification

Logistical Regression

Applying Concepts, Conducting an experiment of tracking coin flip results on a fair coin is an example of

Classification

Applying Concepts, Analyzing patterns and "like" data is an example of _________.

Supervised

Applying Concepts, A learning method based on refining data sets is ________.

Unsupervised

Applying Concepts, A learning method based on gathering data and transforming the data from qualitative context to quantitative context is __________.

-Reconsidering how we communicate and write (e.g., WtBD) -Growth and evolution of the social networking applications

-Development of data literacy

Emerging Trends, The three emerging trends in this chapter presented include (1)____________; (2) ___________, and (3) _________.

Step 1: Outline storyboard (vision casting).

Step 2: Extract and prepare supporting data source(s).

Step 3: Analyze data—Phase 1 (data mining techniques).

Step 4: Incorporate context—Phase 2 (modeling).

Step 5: Populate and display output (data visualization)

Storytelling Placemat steps to tell a story with data

Link Prediction

An attempt to predict a relationship between two data items which may be subjective in. nature. A qualitative approach to apply data mining through technique to a data set. Seen it applied in social media through out network of friends and connections, the content we like or dislike, the schools we attended, etc.

Similarity Matching

A method of identifying similar individuals based on data known about them. Leveraged in fraud detection systems, where we are trying to gather data showing trends and behavior. Any outlined behavior may be suspect for fraud.

Co-occurence Grouping

This method is an attempt to discover associations between individuals based on transactions involving them. Vendors such as Walmart, Target, Amazon, etc. gather as much data about customers bank buying habits as possible. Data is then analyzed on how products might be grouped together, creating marketing strategies to packaged goods together and increase the sale of grouped items.

Casual Modeling

Estimating approach based on the assumption that future value of a variable is a mathematical function of the values of other variables. Used where sufficient historical data is available, and the relationship (correlation) between the dependent variable to be forecasted and associated independent variables is well known. Should not be confused with correlating multiple things together because applying it is used to look at multiple distinct events and identify a pattern where observations can be made where the data supports that one event that seems to be followed by the next event. Understanding the likelihood of event one triggering event two will guide future decision making. Also called causal forecasting model.

-Science

-Art

Storytelling with data is an infusion of _______ with _______. The _______ is selecting the proper methods (e.g., classification, regression, profiling) and building the model (i.e., descriptive, predictive, prescriptive). The _______ is the storytelling component outlined in the storytelling placemat (Figure 2).

Classification

Which approach to Business Analytics attempts to assign each unit in a population into a small set of classes where the unit belongs

Similarity matching

Which approach to Business Analytics attempts to identify similar individuals based on data known about them

Link prediction

Which approach to Business Analytics attempts to predict relationship between two data items

Step 3: Analyze Data- Phase 1

During the module on "Communication - Storytelling" we explored a placemat that provided the frame work for storytelling with data. The step that explores the data mining techniques that we've covered in the class is:

Storytelling Placemat

Which of the placemats, or constructs, that have been provided through Chapter 4 provide the best view of selecting the proper data mining method.

-Context in time is important

-Confirming the business question that needs to be answered

-Refining the answer in iterations as the team builds the analytics solutions

When telling stories, select the major factors that one should consider when building their storyboard: