Version 2:

Instructor: Dr. Liping Liu, College of Business Building 360, 330-972-5947

Office Hours: 3:00-5:00 PM Mondays and Wednesdays

Text Books and References:

• Amit Bahree, Generative AI in Action, Manning, 2024. (ISBN: 978-1633436947)
• A compilation of lecture notes, book chapters, articles, and videos on Artificial Intelligence, Deep Learning, Python, Large Language Models, and Information Systems Fundamentals

Course Description: This course provides a gentle introduction to artificial intelligence (AI) and its applications in business. It offers a broad overview of Generative AI and Large Language Models (LLMs) but focuses on practical prompt engineering and retrieval-augmented generation (RAG) for business productivity. Students will learn how to design effective prompts, work with document collections using RAG, and critically evaluate AI-generated outputs in realistic business contexts.

Design Goal: This course is designed for undergraduate students in all majors to stimulate interest in AI and to promote AI literacy. The emphasis is on helping students become informed professional end users and decision makers in AI-enabled business environments, rather than on advanced programming or statistics.

Prerequisites:  Students are expected to be proficient in using PC and some basic skills of using Excel.

Daily Schedule:

• Week 1: Introduction to AI in Business
  Brief history of AI (symbolic logic at Dartmouth, expert systems, deep learning, generative AI). Overview of AI topics (automated reasoning, image recognition, natural language processing) and business applications (recommendation systems, fraud detection, customer service, logistics, marketing). Discussion: how AI is already affecting students’ daily lives and careers.
  Assignment: Choose a business you know (local store or brand). Write a 2-page brief on how generative AI could be used in marketing, operations, or customer service for this business. Provide one sample prompt you might run today and describe the type of output you expect.
• Week 2: Large Language Models Explained: Neural Networks: introduction to neural networks and deep learning (artificial neurons, layers, feedforward, backpropagation, activation functions: ReLU and Sigmoid). Hands-on lesson 1: Use MIT Netflow to draw neural networks, perform feedforward and backpropagation. Assignment: 1) Create a table to summarize the activation functions, including their function forms, common use cases, and their pros and cons; 2) use Netflow to create a 4-layer neural network with initial weights and one input/output example: input layer: 4 nodes, hidden layer 1: 8 nodes with ReLU, hidden layer 2: 4 nodes with ReLU, output layer: 1 node with Sigmoid. Print the initial network and the final network learned from the one example. 
• Week 3: Large Language Models Explained: Generative AI, Transformers, Generative Adversarial Networks, and LLM Tour
  A non-technical overview of LLMs and a tour of major LLMs: ChatGPT, DALL·E, Gemini, and other LLMs. Explain models as “text and image prediction engines” instead of heavy math. Demonstration of different capabilities: conversation, summarization, translation, and content generation. Introduction to using Google Drive and Colab as platforms to run simple, pre-built AI notebooks without needing programming skills.
  Assignment: Create a simple Colab notebook that contains text cells and code cells (you may use provided template). Include: (1) a short paragraph explaining one AI tool you tried, (2) a code cell that prints a friendly message, and (3) a short reflection on what surprised you the most about generative AI.
• Week 4: Understanding Data and Security
  Analog vs digital data (text, numbers, images, sounds, videos). Basic ideas of digitization and why businesses care about data quality. Introduction to data security concepts: encryption, public vs private keys, hash functions, and why they matter for AI and cloud services. Hands-on: simple examples of turning everyday information into digital data and checking integrity.
  Assignment: Convert a short sentence into ASCII codes (using a table or online tool), and then into binary. In Colab, experiment with simple Python functions like hash() to see how small changes in text change the hash. Write a short paragraph: why is data integrity important when using AI for business decisions?
• Week 5: Prompt Engineering Fundamentals
  In-context learning and prompt techniques: zero-shot, few-shot, and chain-of-thought prompting. Examples of bad vs good prompts for business tasks (customer support replies, report summaries, marketing copy). Live exercises in class: improving prompts step by step and comparing outputs.
  Assignment: Use ChatGPT (or an approved LLM tool) to create your own résumé for a job or internship you might apply for. Submit (1) the final résumé, (2) a list of three prompts you used, and (3) a short reflection on how you improved your prompts to get better output.
• Week 6: From Data to AI – Embeddings, Vector Search, and RAG
  Intuitive explanation of word and document embeddings (turning text into vectors). Concept of vector similarity and how AI “finds similar texts.” Introduction to vector databases and Retrieval-Augmented Generation (RAG): retrieving relevant documents before generating an answer. Business example: searching customer feedback or policy documents.
  Assignment: You will receive a small set of short business documents (e.g., product descriptions or FAQ entries). Using a provided Colab notebook or web tool, ask at least five business questions about these documents via a simple RAG interface. Submit your questions, the AI’s answers, and a short commentary on how accurate and useful the answers were.
• Week 7: Data Models for Business (Tables, JSON, and Graphs)
  Overview of common data models: relational tables, XML, JSON, and graph models. How these models support AI applications and RAG (organizing knowledge for retrieval). Hands-on: represent simple business information in different models (e.g., customer records or product catalogs).
  Assignment: Represent your extended family using (1) a simple table (relational model), (2) JSON, and (3) a graph description (who is related to whom). Include names, birth dates (approximate), jobs, and hobbies. Reflect briefly on which representation you find easiest to read and why.
• Week 8: Midterm Review and Exam
  Review key concepts from Weeks 1–6: AI in business, tool landscape, basic data and security, prompt engineering, embeddings and RAG, and data models. In-class Q&A and practice questions. Midterm exam with both conceptual questions and a short hands-on prompt exercise in AI tools.
  Assignment: No new homework; focus on preparing for the midterm.
• Week 9: Business Analytics with AI – From Numbers to Narratives
  How AI supports analytics: turning raw numbers (from Excel) into plain-language narratives and dashboards. Hands-on: given a simple sales or marketing dataset, use an LLM to generate an executive summary, key insights, and a few recommendations.
  Assignment: Using a provided dataset (e.g., regional sales by month), work in Colab or Excel to (1) create one chart (bar or line) and (2) craft prompts to generate a one-page executive summary. Submit the chart, your prompts, the AI-generated summary, and a short note on what you would change for clarity and credibility.
• Week 10: Prompt Engineering in Colab – Business Workflows
  Using Python code in Colab to automate prompts to ChatGPT or similar APIs (with templates provided). Example 1: send a list of customer reviews and ask the AI to summarize key issues and generate a letter to the Product and Sales departments. Example 2: send a simple financial document and ask for extraction of key figures into structured formats (table/JSON/CSV).
  Assignment: Using a provided Colab notebook, send a small set of customer reviews to an LLM and (1) generate a short summary table of key issues, and (2) draft a short internal memo to management summarizing customer concerns. Submit your prompts, code (template-based), outputs, and a reflection on how helpful and trustworthy the results are.
• Week 11: Prompt Engineering for Data Management
  Business databases: tables, primary keys, foreign keys, simple SQL queries. Use prompts to have an LLM propose database designs and generate SQL commands. Example: tables for customers, orders, and products; ask AI to propose table structures and sample queries. Discuss the importance of checking AI-generated SQL for correctness.
  Assignment: Describe the structure of a small sales database (customers, orders, order lines, products) in plain English. Then ask ChatGPT to (1) propose SQL table definitions and (2) write a query to compute total sales by customer. Submit your description, the AI-generated SQL, and a short critique of what the AI did well or poorly.
• Week 12: Low-Code AI Workflows and Excel Integration
  Explore tools that allow integrating AI into spreadsheets and low-code environments. Example: generate personalized email drafts for customers based on transaction history stored in Excel. Show how prompts can specify output format (tables, lists, bullet points) and how to validate results.
  Assignment: Using a small customer dataset (provided as Excel or CSV), design prompts that ask an AI to (1) segment customers into at least three groups and (2) produce one short email template tailored to each segment. Submit your prompts, AI output, and a brief commentary on how realistic these emails are for actual business use.
• Week 13: Designing AI-Augmented Business Processes and RAG Workflows
  Putting it together: design end-to-end workflows where documents and data feed into prompts and responses. Example: internal FAQ chatbot for HR or IT support using RAG. Students sketch data sources, retrieval steps, prompts, and outputs. 
  Assignment: Using a provided dataset (e.g., regional sales by month), work in Colab or Excel to (1) create one chart (bar or line) and (2) craft prompts to generate a one-page executive summary. Submit the chart, your prompts, the AI-generated summary, and a short note on what you would change for clarity and credibility.
• Week 14: Ethics, Bias, Governance, Privacy, and Risk in Business AI
  Case studies of AI gone wrong: biased hiring systems, privacy violations, misleading recommendations. Discussion of responsible AI principles, corporate governance, and regulatory trends. How prompt design, data selection, and human oversight can reduce risk.
  Assignment: Write a 2-page case analysis of a real or hypothetical AI failure in business (for example, biased resume screening, unfair credit scoring, or misleading product recommendations). Describe the problem, identify where AI and data played a role, and propose at least two mitigation strategies (changed prompts, better data, human review, or policies).
• Week 15: Final Exam
  
  

Alternative Weekly Modules: The above schedule may be adapted with some items be replaced with the following alternatives:

• Prompt Engineering:
  • Prompt Engineering I: The Art of Context (Zero/Few-Shot) Role-Based Prompting: Learn to assign a professional persona ("Act as a Marketing Manager," "Be a Financial Analyst") to get targeted, high-quality output. Understand Tokens as the concept of context length. Assignment (Closeness): The Persona Challenge. Draft a prompt that successfully gets an LLM to generate a full job description for an entry-level position in a field of your choice, demonstrating a clear Role and Task.
  • Prompt Engineering II: Strategic Thinking (Chain-of-Thought) Decision-Making with CoT: Use Chain-of-Thought (CoT) prompting ("Walk me through your reasoning step-by-step...") to force the AI to show its work, improving accuracy and logic for complex business questions. Assignment (Application): Strategic Recommendation. Use CoT prompting to analyze a short case study (provided) on two investment options, requiring the AI to list pros, cons, and a final recommendation.
• Retrieval Augmented Generation (RAG):
  • The RAG Revolution: Connecting AI to Proprietary Data Conceptual RAG: Understand Retrieval-Augmented Generation (RAG). Why it matters: using your company's private documents (e.g., HR manuals, internal reports) to ground the AI's answers, minimizing hallucinations. Assignment (Conceptual Memo): Why RAG? Write a short, non-technical memo (for a CEO) explaining the limitations of a standard ChatGPT query when asking about proprietary company data, and why investing in a RAG system is necessary.
  • Preparing Data for RAG: Chunking & Structuring Data Structuring Basics: Understand that clean, organized data leads to better RAG. Practice transforming unstructured text (e.g., a meeting transcript) into structured formats (outlines, bulleted lists, simple tables in Excel/Sheets). Assignment (Application): Data Standardization. Take a provided, messy internal document (e.g., a policy draft) and use an LLM to extract and format its key rules into a clean, structured list or table, ready for knowledge retrieval.
• Applications:
  • AI for Business Communication and Content Generation Tone and Audience Control: Generate content for different stakeholders: a formal press release (External), a directive memo (Internal), and a catchy social media post. Focus on tone, length, and format control. Assignment (Application): Multi-Channel Campaign. Use an LLM to generate three pieces of marketing content (a formal email, a Twitter/X post, a video script outline) for a single hypothetical new product, demonstrating successful adaptation of tone for each medium.
  • Multimodal AI: Image, Video, and Voice in Business Visual Prompting: Learn to prompt for images using specific modifiers (style, mood, lighting) for marketing materials, website design, and product visualization using tools like DALL-E or Gemini. Assignment (Creative): Brand Identity Prompting. Generate a series of 3-4 images that define the visual brand identity (logo ideas, color palette, mood) for a new, hypothetical startup using an image generation tool.
  • AI for Data Analysis and Interpretation Data Translator: Use LLMs to interpret basic financial or sales data pasted from Excel/Sheets. Prompt the AI to identify trends, calculate simple ratios (without writing formulas), and explain the results to a non-expert audience. Assignment (Analysis): Data Insight Summary. Paste a small sales dataset (provided) into an LLM. Prompt it to identify the top 3 insights (e.g., "highest-selling region," "worst-performing product") and present them as a concise, action-oriented email to the Sales Director.
  • AI and Data Management: The Managerial View Database Concept: What are tables, keys, and schemas? How can AI help us manage them? Use LLMs to draft basic SQL queries or database table definitions based on a natural language request (students critique the AI output, not write code). Assignment (Conceptual): The Database Draft. Write a natural language request (e.g., "I need a database system to track customer orders") and prompt an LLM to generate the corresponding database schema (table names and columns). Critique the AI's design.
  • AI for Business Process Automation Workflow Augmentation: Identify repetitive tasks (e.g., customer service triage, initial contract review, drafting weekly reports). Map out a step-by-step workflow and determine where AI can automate or augment a human's role. Assignment (Strategy): Automation Proposal. Select a repetitive task in a real-world business setting (e.g., HR screening résumés). Write a 1-page proposal on how an LLM could be integrated to streamline this task, including potential benefits (e.g., time saved).
• Project: 
  • AI Business Integration Plan. Choose a major business function (Marketing, Finance, HR) and propose a detailed, non-technical plan for how that department can integrate LLMs (via Prompting, RAG, and/or Multimodal tools) to achieve a specific business goal.
  • RAG for business. Each groups pick a business problem (marketing, operations, strategy) and plan AI-augmented solution (using prompt engineering + RAG). Workshop: students have in-class time to refine problem, identify data/documents, design prompts/workflow, set milestones.

Objectives: This course emphasizes foundational knowledge for AI tools and practical applications of large language models in business. After completing the course, students should be able to:

• Understand key concepts for applying large language models (LLMs), such as prompts, tokens, neural networks, transformers, generative adversarial networks, learning, and evaluation
• Understand the data engineering process at a high level, including analog and digital data, data security and privacy, structured vs. unstructured data, and data models
• Develop zero-shot, few-shot, and chain-of-thought prompts for popular LLMs to generate text, images, and simple code for data management and analytics.
• Explain, at a high level, how embeddings, vector databases, and retrieval-augmented generation (RAG) support business applications.
• Design simple AI-augmented business workflows using prompts, documents, and spreadsheets, including internal FAQ chatbots and analytics summaries.
• Critically assess ethical, governance, privacy, and risk issues related to AI deployment in organizations.

Exams: This course will have two major exams as scheduled above. Each exam includes both hands-on and written problems.

Assignments: Homework is assigned once a week for 12 weeks; each consists of conceptual questions and hands-on projects classified into three grading categories: correctness, closeness, and completeness. The correctness problems will be graded by ecourse.org, and closeness questions are graded and/or commented by instructors. Students will earn points automatically for each completeness question if it is finished (it has to be deemed complete). Assignments are due at the beginning of classes meetings on Mondays (except for holidays). No late homework will be graded. Please show your work in a neat and orderly fashion. Write or type your work on one side and in every other line. Use standard size paper (8 1/2'' by 11''). Do not use spiral notebook paper.

Attendance: Attendance is MUST and will be 10% of your final grade. Attendance will be managed by ecourse.org. The formula for computing your attendance grade is non-linear. It will take one point off for the first absence, 2 points off the second, 3 points off the third, and 4 points off the fourth. If you missed the equivalent of three-week classes, you fail the course automatically. Under special situations, you can take some classes online with the following guidelines:

1. You must obtain permission from the instructor at least one day ahead of each online session
2. Follow the lectures or recordings to perform all in-class hands-on exercises and take notes. Within one day from the class submit your notes and the finished exercises to ecourse.org as Proof of Attendance.
3. All weekly assignments are due at the same time as in-person classes. All exams must be onsite.

Quizzes: I will use quizzes regularly to check your completion or preparation of assignments

Makeup: Each student with appropriate excuses may have at most one chance to makeup homework or quiz. Note that it is your privilege but not your right to have this special favor. Also, all makeups must be completed within one week of due date and before answer key is released. 

Grades: Your final grades will be calculated by the following formulas:

35% (HW) + 55% (Tests) + 10% (Attendance)

A = 93-100%; A– = 90-92%; B+ = 87-89%; B = 83-86%; B– = 80-82%; C+ = 77-79%; C = 73-76%; C– =70-72%; D = 60-69%; F = 59% and less

Misconduct: Academic misconduct by a student shall include, but not limited to: disruption of classes, giving and receiving unauthorized aid on exams or in the preparation of assignments, unauthorized removal of materials from the library, or knowingly misrepresenting the source of any academic work. Academic misconduct by an instructor shall include, but not limited to: grading student work by criteria other than academic performance or repeated and willful neglect in the discharge of duly assigned academic duties. Convicted violations may result in grade penalties, besides the school official ones, such as increased scrutiny of future submissions, reduced benefits of curving, if any, and/or the reduction of overall grade. 

On Collaboration: All for-credit assignments, except for those designated as group projects, must be done independently, and collaboration in providing or asking for answers to those assignments constitutes cheating. 

On AI Tools: In this class, I allow students to use AI tools to help their learning. However, submitting AI generated work in lieu of your own personal contribution for credits is a violation of academic code. If a submitted work is suspected to be AI generated, the student will be asked to reproduce the submitted work in front of the instructor. 

Looking  for additional help? Students looking for additional assistance outside of the classroom are advised to consider working with a peer tutor through Knack. The University of Akron CBA has partnered with Knack to provide students with access to verified peer tutors who have previously aced this course. To view available tutors, visit uakron.joinknack.com and sign in with your student account. At the same time, if you are doing well in this class, please go to uakron.joinknack.com where you can create a verified tutoring profile and begin helping other students.

Version 1:

Instructor: Dr. Liping Liu, College of Business Building 360, 330-972-5947

Office Hours: 3:00-5:00 PM Mondays and Wednesdays

Text Books and References:

• Amit Bahree, Generative AI in Action, Manning, 2024. (ISBN: 978-1633436947)
• A compilation of lecture notes, book chapters, articles, and videos on Artificial Intelligence, Deep Learning, Python, Large Language Models, and Information Systems Fundamentals

Course Description: This course aims at a gentle introduction to artificial intelligence (AI) and its applications. It provides a broad overview of the core concepts of Generative AI and Large language models but focuses on prompt engineering and retrieval-augmented generation (RAG) with LLMs for business productivity. It is designed to stimulate interests and lead students to be literate on AI to become professional end users and decision makers in AI integration and applications.

Design Goal: This course is designed for all undergraduate students, irrespective their major areas, to stimulate their interests in and to promote their understanding about AI. 

Prerequisites:  Students are expected to be proficient in using PC and some basic skills of using Excel.

Daily Schedule:

• Week 1: Introduction (Brief History of AI: symbolic logic at Dartmouth, expert systems, deep learning, and generative AI, and the story on the auction of Geoffrey Hinton. Overview of AI Topics: automated reasoning and planning, theorem proving, knowledge representation, image recognition, natural language processing. Overview of AI methods: symbolic logics, artificial neural networks, genetic programming, and Bayesian inference. AI applications and success stories: robots, chess playing, self driving, fraud detection, resource scheduling, etc. Assignment: Create an account with OpenAI for using ChatGPT, and create an account with Google and setup google Drive and use Colab and Gemini. 
• Week 2: A tour of LLMs: ChatGPT, DALL-E, Sora, Grok, LIama, Deepseek, and Gemini. Hands-on: set Google Drive and Colab, and use Colab to create notebooks with text and code for numerical calculations and load Python modules: import math for using math.cos(), math.exp(), etc. Assignment: Use Colab to create a notebook with text and code cells for the following tasks: 1. convert temperature of 78 Fahrenheit to Celsius, 2. compute the area of a circle of diameter 28 meters; 3. compute the total number of seconds of the time 13:34:51 passed from the midnight at 0:0:0, (check for completion)
• Week 3: Data Digitization and Security: Analog Data Types: text, numbers, images, sounds, videos. Data Digitization, Data Security: cryptography, symmetric encryption, public and private keys, and hash functions. Hands-on lesson 1: how to digitize texts, images, sounds, and videos. Hands-on Lesson 2: use Colab: use hash() function in Colab and from cryptography.fernet import Fernet to practice symmetric encryption and decryption. Assignment: 1) Use ASCII table to convert this sentence into binary numbers: "AI is fun." 2. use bin() function in Colab convert a decimal integers 23456 and 98076 into a binary numbers; 3) Write any secret message and use cryptography.fernet to generate a key and use the key to encrypt the message to give to your friend and ask your friend to use the key to decrypt the message.  
• Week 4: Prompt Engineering (in-context learning and prompt techniques: zero-shot, few-shot, chain-of-thought prompting). Exercise 1: use CoT technique to prompt ChatGPT for solving homework questions, translate languages, and write a literature review on a topic. Exercise 2: use DALL-E to generate arts and product designs with specific art styles and modifiers in image quality, lighting, situations, etc. Assignment: Use ChatGPT to create your own resume for a job application and submit a list of your prompts and responses as a Word document (check for completion).
• Week 5: Word Embedding, Vector Data Database, and Retrieval-Augmented Generation (RAG). Hands-on Exercise 1: use tiktoken for tokenization; Exercise 2: use openai for word embedding; Exercise 3: use sapCy for chunking enterprise documents for RAG. Assignment: Tokenize a paragraph of about 50 words, obtain word embeddings, and find which words are closed in meaning.
• Week 6: Data Models: XML, JSON, relational model, and graph model. Hands-on Lesson: organize Amazon reviews into a list, and organize two product data using JSON. Assignment: Create a XML file, a JSON file, a graph model, and relational model to represent all members in your extended family and their relationships, including details like their names, birth dates, jobs, and hobbies.
• Week 7: Midterm Exam
• Week 8: Prompt Inputs and Outputs: prompt with multiple curated examples and data points, prompt with required output data models,  Data structures (python data types: strings, numbers, and date/times, lists, and dictionary, for-loop control to process lists, and if-else decision control to process list items selectively.) Assignment: Create lists of student ages and another list of their part-time work earnings, and compute the total amount of earnings and the average amount of earnings by students who are age 20 or over. (check for correctness)
• Week 9: Prompt Engineering in Colab (Use Python code in Colab to query ChatGPT: Exercise 1: send a list of customer reviews on a product,  ask ChatGPT to create a summary of the reviews into an HTML table, JSON data objects, and an Excel spreadsheet on the issues mentioned in the reviews, and write a letter to the Production Department and another letter to Sales Department for actions addressing the issues. Exercise 2: send a financial document and ask ChatGPT to extract financial information and output the result as HTML table, JSON data objects, and an Excel spreadsheet.
• Week 10: Prompt Engineering for Data Management (database, primary keys, and foreign keys, SQL queries, specify table names and the columns of each table for human resource management and ask ChatGPT to generate SQL commands to create tables, insert sample data, and retrieve information in Microsoft Access, mySQL, and Oracle. Assignment:  specify table structures of a sales databases including customers, orders, orderlines, and products. Then ask ChatGPT to create SQL commands for MS Access to create the tables, insert sample data, and retrieve the total amount of orders made by each customer. (Check for completion)
• Week 11: Prompt Engineering for Excel Modeling (Exercise 1: load data from Excel into Colab and summarize and visualize data. Exercise 2: Use ChatGPT to generate python code and modify and run code for creating an Excel spreadsheets for products including their description, sizes, colors, in-stock quantities, and prices, generate code to compute total inventory values, compute the number of products which is under-stocked. Assignment: Use ChatGPT to generate code to create an Excel spreadsheet to manage personal daily expenses, compute total expenses every month, and plot the trend of monthly spendings. (check for completion)
• Week 12: Prompt Engineering for coding: Python Coding Basics, turn python lists into numpy arrays and pandas time series and data frames, load Excel data into data frames. Assignment: create five lists of student names, birth dates, majors, admission dates, part-time work earnings. Then create a data frame using the list, save the data frame as an Excel file, and load the file back to Colab. (check for completion)
• Week 13: Artificial Intelligence: introduction to neural networks and deep learning (artificial neurons, layers, feedforward, backpropagation, activation functions: ReLU and Sigmoid). Hands-on lesson 1: use Colab for feedforward computation of 3-layer neural networks using ReLU and Sigmoid activation. Hands-on Lesson 2: Use MIT Netflow to draw neural networks, perform feedforward and backpropagation. Assignment: 1) Create a table to summarize the activation functions, including their function forms, common use cases, and their pros and cons; 2) use Netflow to create a 4-layer neural network with initial weights and one input/output example: input layer: 4 nodes, hidden layer 1: 8 nodes with ReLU, hidden layer 2: 4 nodes with ReLU, output layer: 1 node with Sigmoid. Print the initial network and the final network learned from the one example. 
• Week 14: Artificial Intelligence: create neural networks with TensorFlow for linear regression (keras and tensor flows tools, dense layers, input shapes, loss, metrics, optimizers, training and evaluation) Assignment: predict insurance claim costs using a network of four layers of 64 neurons with relu activation, 32 neuros with relu activation, dropout with rate = 0.5, and output layer of one neuron with linear activation. 
• Week 15: Final Exam

Objectives: This course emphasizes the foundational knowledge for artificial intelligence (AI) tools and applications of large language models for business. After the course, the student should be able to

• Understand key concepts for applying large language models (LLMs) such as prompts, probabilities, temperatures, tokens, data, training, evaluation, modality, etc.
• Articulate foundational technologies behind generative AI including neural networks, transformers, and generative adversarial networks.
• Understand issues and solutions for corporate integration of large language models
• Develop zero-short, few-short, and chain-of-thought prompts for popular large language models to generate audio, video, art, texts, and code for data management, spreadsheet modeling, information security, and data analytics
• Apply generative AI techniques to business integration including tokenization, word embedding, vector databases,  retrieval augmentation generation, prompt engineering, etc.
• Use Google Colab and Python code to query large language models. 

Exams: This course will have two major exams as scheduled above. Each exam includes both hands-on and written problems.

Assignments: Homework is assigned once a week for 12 weeks; each consists of conceptual questions and hands-on projects classified into three grading categories: correctness, closeness, and completeness. The correctness problems will be graded by ecourse.org, and closeness questions are graded and/or commented by instructors. Students will earn points automatically for each completeness question if it is finished (it has to be deemed complete). Assignments are due at the beginning of classes meetings on Mondays (except for holidays). No late homework will be graded. Please show your work in a neat and orderly fashion. Write or type your work on one side and in every other line. Use standard size paper (8 1/2'' by 11''). Do not use spiral notebook paper.

Attendance: Attendance is MUST and will be 10% of your final grade. Attendance will be managed by ecourse.org. The formula for computing your attendance grade is non-linear. It will take one point off for the first absence, 2 points off the second, 3 points off the third, and 4 points off the fourth. If you missed the equivalent of three-week classes, you fail the course automatically. Under special situations, you can take some classes online with the following guidelines:

1. You must obtain permission from the instructor at least one day ahead of each online session
2. Follow the lectures or recordings to perform all in-class hands-on exercises and take notes. Within one day from the class submit your notes and the finished exercises to ecourse.org as Proof of Attendance.
3. All weekly assignments are due at the same time as in-person classes. All exams must be onsite.

Quizzes: I will use quizzes regularly to check your completion or preparation of assignments

Makeup: Each student with appropriate excuses may have at most one chance to makeup homework or quiz. Note that it is your privilege but not your right to have this special favor. Also, all makeups must be completed within one week of due date and before answer key is released. 

Grades: Your final grades will be calculated by the following formulas:

35% (HW) + 55% (Tests) + 10% (Attendance)

A = 93-100%; A– = 90-92%; B+ = 87-89%; B = 83-86%; B– = 80-82%; C+ = 77-79%; C = 73-76%; C– =70-72%; D = 60-69%; F = 59% and less

Misconduct: Academic misconduct by a student shall include, but not limited to: disruption of classes, giving and receiving unauthorized aid on exams or in the preparation of assignments, unauthorized removal of materials from the library, or knowingly misrepresenting the source of any academic work. Academic misconduct by an instructor shall include, but not limited to: grading student work by criteria other than academic performance or repeated and willful neglect in the discharge of duly assigned academic duties. Convicted violations may result in grade penalties, besides the school official ones, such as increased scrutiny of future submissions, reduced benefits of curving, if any, and/or the reduction of overall grade. 

On Collaboration: All for-credit assignments, except for those designated as group projects, must be done independently, and collaboration in providing or asking for answers to those assignments constitutes cheating. 

On AI Tools: In this class, I allow students to use AI tools to help their learning. However, submitting AI generated work in lieu of your own personal contribution for credits is a violation of academic code. If a submitted work is suspected to be AI generated, the student will be asked to reproduce the submitted work in front of the instructor. 

Looking  for additional help? Students looking for additional assistance outside of the classroom are advised to consider working with a peer tutor through Knack. The University of Akron CBA has partnered with Knack to provide students with access to verified peer tutors who have previously aced this course. To view available tutors, visit uakron.joinknack.com and sign in with your student account. At the same time, if you are doing well in this class, please go to uakron.joinknack.com where you can create a verified tutoring profile and begin helping other students.