IDSC - Test 4

Accuracy of Personal Data

- Alice Arias story on subsidized housing - Fair Credit Reporting Act (FCRA) allows data subjects to sue a credit bureau for damages if the bureau has been notified repeatedly about errors but has taken no suitable steps to correct them

Advantages of the Waterfall Model

- Allow the project as a whole to maintain a more detailed, robust scope and design structure due to all the upfront planning and documentation stages. - Forces structured organization and be disciplined in the design and structure - Well-suited for organizations or teams that work well under a milestone- and date-focused paradigm

Potential Disruptor Spotted: Now What?

- Build a portfolio of options on emerging technologies, investing in firms, startups, or internal efforts - Options give the firm the right to continue and increase funding as a technology shows promise - Encourage new market and technology development

Bitcoin Concerns

- Consumer benefit needs to be stronger - Difficult to understand/use technology - Bad reputation (drug dealers, tax evaders, etc.) - Firms have struggled under an ambiguous cloud of not knowing how they will be regulated and what legal issues apply - Volatility of the value - Still some security issues

Collection and Storage of Personal Data

- Don't ask more data than subjects believe is really needed (Radio Shack discontinued in late 2002 its long-time practice of asking customer for their phone number, name, address) - Explain clearly how the data will be used - Offer options to provide part of the data ("not take it or leave it")

Authorized Access to Personal Data

- Need technical controls to allow restricted access to sensitive data - Create policies to describe who does and does not have such a "need to know" - Physical blocking devices in clicks - People get fired for unauthorized access

Waterfall model phases

- Requirements - Analysis - Design - Coding - Testing - Operations

Benefits of Cryptocurrency

- eliminates transaction fees - possibilities of micropayments (or small digital payments) that are now impractical because of fees - could be a boon for international commerce, especially for cross-border remittance and in expanding e-commerce in emerging markets - blockchains have no single controlling entity or intermediary where fraud, corruption, damage, hacking, or government shutdown could occur (hence more secure)

Six Managerial Decisions

-Data collection and storage -Secondary data use -Data accuracy -Authorized access -Automated judgment -Data merging and profiling

Software Development Processes

1. Requirements Analysis 2. Design 3. Coding & Testing 4. Deployment & Maintenance

Daily Scrum

15 minute time boxed meeting held at the same time, every day of the Sprint, where the previous day's achievements are discussed, as well as the expectations for the following one.

Food Supply Keychain

A consortium with 10 major food suppliers will apply to block chain to improve food safety and ingredient transparency Blockchain can help brands track sources of contamination far more quickly reducing the impact of compromised foods.

Cryptocurrencies

A digital asset where a secure form of mathematics is used to handle transactions, control the creation of additional units, and verify the transfer of assets

Blockchain Technology

A distributed and decentralized ledger that records and verifies transactions and ownership - key words: distributed and decentralized

The Waterfall Model

A linear or sequential development process, in which development is seen as flowing steadily downwards

Sprint backlog

A list of the tasks and requirements that need to be accomplished during the next Sprint. Sometimes accompanied by a Scrum task board, which is used to visualize the progress of the tasks in the current Sprint, and any changes that are made in a 'To Do, Doing, and Done' format.

Sprint retrospective

A meeting where the Scrum team reflect on the proceedings of the previous Sprint and establish improvements for the next Sprint.

Product vision statement

A summary that articulates the goals for the product.

Systems Development

A systems or software development process is a structure imposed on the development of an information systems solution, to organize the activities that go into producing the solution to an organizational problem or opportunity The process of creating and maintaining information systems. Developing an information system can involve all five components: hardware, software, data, procedures, and people.

Release plan

A timetable for the release of a working product.

User stories

A tool used to explain a software feature from an end-user perspective, a simplified description of requirements (who needs what and why) As a [role], I want [feature], because [reason].

Kanban swimlanes

A visual element on the board that is used to distinguish tasks/items by categorizing them.

Kanban board

A visual management tool used to visualize the development process.

Neural Networks

AI algorithms that recognize underlying relationships in a set of data through a process that mimics the way the human brain operates (biological neural networks)

Agile Development

Agile software development refers to software development methodologies centered round iterative development, where requirements and solutions evolve through collaboration between self-organizing cross-functional teams

Examples of Customer Insights

Algorithms to detect patterns in vast volumes of data and interpret the meaning ◦Predict customer purchase or churn ◦Identify credit fraud in real time ◦Analyze warranty data to identify safety issues ◦Automate personalized targeting of digital ads

Sprint review

An informal meeting held at the end of every Sprint where the Scrum team present their Increment to the stakeholders, and discuss feedback.

Bitcoin

An open source, decentralized payment system that operates in a peer-to-peer environment, without bank or central authority

Generative AI

Artificial intelligence that can produce content (text, images, audio, video, etc.) such as ChatGPT. It operates similarly to the "type ahead" feature on smartphones that makes next-word suggestions. Gen AI is more sophisticated and based on the particular content it was trained on (exposed to)

Examples of Process Automation

Automation of physical and digital tasks ◦Transfer data from email and calls to other systems ◦Replacing lost credit or debit cards ◦"Reading" legal and contractual documents to extract provisions using NLP

Examples of Disruptive Techs

Blockchain Artificial intelligence Virtual Reality / Augmented Reality Digital cameras iPod, iPad, iPhone E-commerce models (Amazon) Sharing economy (Uber & Airbnb) Massive Online Open Course (MOOC)

Strong AI

Can do most everything we humans can do and possibly much more

Weak AI

Can perform only one narrowly defined task or a small set of related tasks

Hyperautomation

Can provide global connectivity across the company and create a complete system that (automatically) synchronizes all departments

Kanban cards

Cards that depict a work item/task in the work process

AI Bias in Face Recognition

Commercial face-recognition systems tend to be more accurate on white male faces than on female or nonwhite faces.

IoT Enabling Technologies

Communications networks and protocols, hardware devices and components such as sensors, wireless charging technologies and software.

Artificial intelligence

Computer software that can mimic or improve upon functions that would otherwise require human intelligence

Another Set of IoT Applications

Consumer and Residential ◦Remote monitoring and control of appliances, lighting, heating Health and Healthcare ◦Wrist bands and wearable devices Transportation ◦Smart cars, smart roadways to minimize congestion Energy Infrastructure and Distribution ◦Smart grid to estimate usage and distribute energy Public Safety ◦Smart apps to pay parking or to report potholes Industrial and Manufacturing The Environment

Actuators

Convert energy to motions.

Challenges with AI Projects

Data quality issues & inconsistent data Not enough training data (rare events) AI skills are rare (hundreds of people) Lack of transparency or explainability Need for change management

Deep Learning

Deep neural networks have more than one layer of hidden units (Its depth is the number of hidden layers) Deep learning is a class of machine learning algorithms in the form of a neutral network that uses more than one hidden layers of processing units to extract features about new data.

Blockchain - How Does It Work?

Each block on the chain contains a number of transactions and every time a new transactions occurs, a record of the transaction is added to each network participant's ledger. Trust is established NOT by intermediaries, but by mass collaboration and clever code, which makes it difficult to change or cheat the system.

Why Big Firms Fail

Failure to see disruptive innovations as a threat. - Reason: They do not dedicate resources to developing the potential technology since these markets do not look attractive. - Creates blindness by an otherwise rational focus on customer demands and financial performance. Start-ups amass expertise quickly. - Big firms are forced to play catch-up. - Few ever close the gap with the new leaders

Development team

Group of professionals who deliver the product (developers, programmers, designers).

Different Cultures on Privacy

Human rights countries - View privacy as human rights - "Omnibus" privacy bills - Federal privacy bureau with proactive power Contract term countries - View privacy as contractual negotiations - "Sectoral" laws (legal liabilities defined for medical, credit card, video rental records) - Rely on self-help and voluntary control models

4 Agile values

Individuals and interactions over processes and tools Working software over comprehensive documentation Customer collaboration over contract negotiation Responding to change over following a plan

sprint

Iterative time boxes where a goal is accomplished. Time frame does not exceed one calendar month and are consistent throughout the development process.

Scrum goal

Its goal is to develop, deliver, and sustain complex products through collaboration, accountability, and iterative progress.

Kanban

Kanban is a highly visual method popularly used within agile project management.

Continuous integration

Keeping the code up to date by producing a clean build of the system few times per day, which enables the delivery of a release at any moment

Sensors

Low-powered, wireless microelectronic devices that are designed to monitor a specific physical, electrical or chemical element and to transmit data on changes in that element over time. - Key words: Monitor and communicate

Product backlog

Managed by the Product Owner, it's where all the requirements needed for a viable product are listed in order of priority. Includes features, functions, requirements, enhancements, and fixes.

Non-Fungible Tokens (NFTs)

NFTs are a type of digital asset that use blockchain to document the ownership of items such as images, videos and other collectibles. "Non-fungible" essentially means unique - the digital version of an original piece of artwork. They are stored on the blockchain; as the blockchain is transparent, it's easy for everyone to see who owns which token.

Examples of Customer Engagement

NLP chatbots and conversational agents ◦Intelligent agents offer 24/7 customer service ◦Answer employee questions on IT / HR ◦Product and service recommendations ◦Health treatment recommendations

Examples of Blockchain in Action

Nasdaq stock exchange The world's leading financial institutions, including Bank of America, Citigroup, HSBC, ING IBM Insurer Nationwide Overstock.com De Beers Spotify Kodak Facebook

Disadvantages of the Waterfall Model

Non-adaptive design constraints - Inherent lack of adaptability across all stages of the development life cycles Ignores mid-process user/client feedback - User or client feedback provided late into the development cycle can often be too little, too late Delayed testing period - Most bugs or even design issues won't be discovered until very late into the process

Opt-In or Opt-Out?

Opt-in: unless the data subject has given an overt permission for some future use of the data, it must be assumed that there is no consent for that use. Opt-out: unless the data subject registers an overt objection, it is assumed that the additional use of the data is acceptable.

Product backlog

Ordered by priority, this is the full list of what is needed to be done to complete your project.

Scrum master

Organized servant-leader who ensures the understanding and execution of Scrum is followed

Three Types of AI Applications

Process automation or RPA ◦Automation of physical and digital tasks Customer insights ◦Algorithms to detect patterns in vast volumes of data and interpret the meaning Customer engagement ◦NLP chatbots and intelligent agents

Scrum Artifacts

Product Backlog, Sprint Backlog

Product Owner

Product expert who represents the stakeholders and is the voice of the customer.

Scrum Team Roles

Product owner Development team Scrum master

Agile Deliverables

Product vision statement, product roadmap, product backlog, sprint backlog, release plan, increment

Project Management Certifications

Project Management Professional (PMP) Certified Associate in Project Management (CAPM) Project+ Professional

Job Placement

PwC predicted that between 22% and 44% of today's current jobs could be automated by the 2030s.

Reinforcement Learning

Requires no labeled training examples Instead, an agent - the learning program - performs actions in an environment (usually a computer simulation) and occasionally receives rewards from the environment

Scrum

Scrum is an agile framework that is used to implement the ideas behind agile software development.

Internet of things challenge

Security being the greatest challenge

Secondary Use of Personal Data

Significant restriction on secondary use of medical data in the U.S. External versus Internal secondary data use (car dealership example) Concerns with affiliate sharing "creative" use of friend data on Facebook

Is the Algorithm Racist or Sexist?

Since AI systems "learn" based on human-labeled data, then any biases in data can become part of the model. Implications of such systems range from poor customer service to harmful, incarceration-causing misclassification by criminal justice and security systems.

IT Applications

Smart Home Systems ◦Remote light control ◦Thermostat control ◦Security cameras ◦Niwa - an automated watering system Manufacturing ◦Engine sensors to reduce wear ◦Heat and humidity censor of plants ◦Automatic order of replacement parts for machines Retail ◦Automated supermarket Amazon Go ◦Smart cart by Panasonic Restaurants ◦Automatic product ordering ◦Expiration date tracking Agriculture ◦Track movements and health of cattle ◦Precise farming by analyzing soil and determine amount of water and fertilizer

Scrum Events

Sprint Sprint Planning Daily Scrum Sprint Review Sprint Retrospective

The Theory of Stakeholder

Stakeholders are anyone with a stake or claim on the firm such as stockholders, customers, employees, suppliers, and the local community - Respect the rights of each stakeholder - Give equal consideration to the legitimate interests of all stakeholders and to adopt corporate policies that produce the optimal balance among them

Steps to Take When Quandaries Arise

Stockholder Perspective - Review legislation and administrative mandates - Avoid deception and fraud - Assign probabilities to the possible outcomes from alternative actions Stakeholder Perspective - Identify all stakeholders - Determine each stakeholder's right - Consider each stakeholder's right

Machine Learning Approaches

Supervised learning: Learning with a labeled set Unsupervised learning: Discovering patterns in unlabeled data Reinforcement learning: Learning based on feedback or reward

What is algorithmic bias

Systematic and repeatable errors that create unfair outcomes, such as privileging one arbitrary group of users over others. For example, a credit score algorithm may deny a loan without being unfair, if it is consistently weighing relevant financial criteria. (Wikipedia.com)

Expert Systems

Systems with explicitly written rules or examples to perform a task in a way that mimics human expertise. ◦Explicitly write the rules for the machines ◦Versus having machines learn on their own in Machine Learning

Disruptive Technologies

Technologies that create market shocks and catalyze growth (a.k.a. disruptive innovation). They have two characteristics that make them threatening. ◦First, they come to market with a set of performance attributes that existing customers don't value. ◦Second, over time the performance attributes improve to the point where they invade established markets.

Prototyping Model

The activity of creating prototypes of software applications, i.e., incomplete versions of the software program Used to visualize some component to limit misunderstanding of requirements

Product roadmap

The high-level view of the requirements needed to achieve the product vision.

The Theory of Stockholder

The managerial obligation is to maximize the financial returns of the stockholders, with two constraints and one assumption: - Through legal, non-deceptive means - To have a long-term orientation - Pursuing profits also promotes the interests of the society

Internet Of Things (IoT)

The multiple networks of devices or technology platforms ("things") that communicate with each other via wireless protocols and without direct human interaction (e.g., smart home devices like programmable thermostats) - Can lead to increased efficiency, better performance and enhanced safety

Key distinction between stockholder and stakeholder

The stockholder theory sees all parties except stockholders as means to the end of profitability. The stakeholder theory sees all stakeholders as ends in themselves.

Sprint backlog

The user stories (requirements), goals, and tasks linked to the current sprint.

Increment

The working product functionality that is presented to the stakeholders at the end of the sprint, and could potentially be given to the customer.

Blockbuster Video Case

To categorize its 30 million customers and to sell information to direct mailers for target-marketing campaigns According to the stockholder theory Legal? Deceptive? Profitability? According to the stakeholder theory Stakeholders? Rights and Interests

Health Care Application - Medical records

Transactions are 'granting of access rights' to specific healthcare providers. Only the patient knows their private key, thus only the patient can grant these rights. Transactions are verified by miner nodes (operated by healthcare organizations), which, in turn, earn access rights to (aggregated / anonymized) patient data, e.g., aggregation of data tied to blocks in the chain

How Does Bitcoin Work?

Transferred from person to person like cash Transactions are recorded in a distributed, decentralized public ledger, known as a blockchain Verification and time-stamping of transactions is performed by a pool of users called miners No one can transfer the asset without a special password called a private key, usually stored in a cryptocurrency wallet, which is really just an encrypted holding place Passwords are virtually impossible to guess, and verification makes sure no one spends the same bitcoins in two places at once

Scrum distinguishment

What distinguishes Scrum from other agile methodologies are the roles, events, and artifacts that it is made up of, with which it uses to operate.

sprint planning

Where the entire Scrum team get together — at the beginning of every Sprint — to plan the upcoming sprint.

AI Bias in Word Association

Word associations based on Google News articles capture biases inherent in the language data that produces them

Customer collaboration

communication with customers to keep them aware and provide feedback

Agile promotes:

iterative development, team collaboration, and change recognition

Transparent / Explainable / Ethical AI

the "black box" nature of machine learning and deep neural networks, with multi-layered statistical weightings, makes it difficult to understand how the systems make their decisions - the possibility of unintended consequences - earlier this century, the financial industry used derivative models so complex that they eventually contributed to a global recession-causing crisis

Large Language Models (LLM)

◦Multiple layered neural networks trained on massive text data sources, to understand existing content and generate original content

Prompt Engineering

◦Process of crafting effective prompts to guide language models (e.g., llama playing basketball)

Machine learning

◦Software that contains the ability to learn or improve through experience without being explicitly programmed. ◦Algorithms that find patterns in data ◦It is a subset of Artificial Intelligence