COGS 109: Modeling and Data Analysis (Summer 2017)

Course website:


Megan Bardolph (Professor)

Office Hours: MW 10am in CSB 237

Carson Miller Rigoli (TA)

Course Description

Welcome to COGS 109!

This course will introduce you to computational methods used for exploring and characterizing data, creating and training models to predict data, and comparing these models. You will primarily use MATLAB to implement these models and perform analysis techniques. We will focus on probability throughout the course as a framework for evaluating models. Students will practice communicating their knowledge and their results by curating content online and presenting during class.

Class time will be highly interactive, consisting of a mixture of individual and group work, iClicker polling, large class discussion, and lecture time. Attendance and participation are important and will be part of student grades.


iClickers will be used in class and lab to enhance engagement on a daily basis. Clickers can be purchased at the bookstore. They must be registered for the class through the TED website.

There are no reading materials to be purchased.

Learning objectives

By the end of this course, you will be able to:

  • Apply data analysis and modeling techniques.
  • Explain principles of data science using both plain English and proper terminology.
  • Select and defend a given technique based on the dataset and objective.
  • Use MATLAB to manipulate and analyze data.
  • Visualize, interpret, and confidently communicate results of data analyses.
  • Evaluate data analysis projects by assessing appropriateness of technique, success of technique, and whether the research question was answered.

Course Details and Policies

Course schedule, content, and announcements:

Class and homework discussion:

Homework submission and grading, file distribution:

Contact information

I encourage students to attend office hours with me (the instructor) and the TA to clarify concepts and ask questions. If you need to schedule meeting time outside of office hours, please ask in person before or after class, or email to set up a time. Please expect up to a 24-hour wait time for an email response. Questions about grading, syllabus, and content should NOT be emailed – please ask on Piazza so the whole class can benefit from the discussion.

Classroom policies

During class time, we will use active learning techniques along with traditional lecture. This means students will ask and answer questions in class, collaborate with other students, and move around the room. If for any reason an activity makes you uncomfortable, please let me know so I can help you find an alternative way to participate.

Please be considerate of your classmates and avoid activities that distract from learning, such as texting, web browsing, and chatting during lecture. You are encouraged to actively participate and engage with other students during active portions of each class.

Attendance will be taken during every class meeting and will count toward your grade. Full points are earned by attending 80% of classes, meaning you have up to 2 excused absences. Attendance may be taken at any time, including the beginning and end of class. Attendance will be mandatory for some sections, particularly toward the end of the course, when you will be working on group projects.

Diversity is present in every class because students come from different backgrounds and bring different knowledge and experiences with them. I expect every student to be respectful of differences and I aim to create a welcoming, inclusive environment where every student has the resources they need to learn. Generating content (blog posts, etc.) that is derogatory toward another person or group will not be tolerated.

Grading policy

Percentages assigned to each item were chosen by students during class. Please see Syllabus document for details (on Tritoned and Piazza).

You must take the midterm exam, complete the final project, and attempt every assignment to receive a passing grade.

Late work policy

Late work will not be accepted, except in the case of a documented short-term illness or emergency. If this is the case, please contact the course instructor as soon as possible.

Regrade policy

If you believe there has been an error in grading your assignment, please submit a re-grade request to the instructor within three days of the assignment being returned. Submit your assignment along with a cover sheet explaining the error.

If you think your work deserves more points (different from correcting a numerical error), please include in your cover sheet a concise description of how your answer compares to the rubric and why you think it should have earned more points. Please understand that your entire assignment may be re-graded, meaning your overall grade may go up, down, or remain the same.

Regrade requests for the midterm exam must be submitted in class on the same day you receive the graded exam.

Academic integrity

I expect all students to honor the highest standards of academic integrity in this course. Please read and be familiar with the UCSD Policy on Integrity of Scholarship (

Overall, this means that all academic work is done by the individual who takes credit for the work, without unauthorized aid of any kind. For homework, this means you can work together, but you must write your own code without looking at that of others. For example: you could ask someone for help and discuss the code needed to accomplish your goal. You cannot, however, take notes of theirs and then immediately write the same thing. You must sit down and write your own code on your own, so that you know that you have processed what you have learned and can produce the necessary code on your own.

The teaching team may check submitted assignments, including code, for plagiarism, either by comparing assignments individually or using automatic detection software. If you have any questions about what types of actions are or are not allowed, please discuss them with someone from the teaching team before taking the actions.

Course Schedule

Week Date Topic Materials Assignments Notes
1 Mon Jul 03 Intro/Matlab review Lecture 1 handout:
Lecture 1 slides:
Matlab HW

Wed Jul 05 Probability
Probability HW
2 Mon Jul 10 Linear regression

Wed Jul 12 PCA

3 Mon Jul 17


Wed Jul 19

Clustering / Technique selection

4 Mon Jul 24 Midterm / Probability

Wed Jul 26 Project work

5 Mon Jul 31 Perceptron/neural nets

Wed Aug 02 Project work/neural nets

Fri Aug 04 Final exam period: Project presentations