Invest in our country's and our region's technological future. Gain more than 900 hours of focused attention for your project.
What is a Capstone? Who are Capstone Sponsors? Who are Capstone Students?
Benefits to the Sponsor Suitable Project Concepts Expectations for Capstone Sponsors
The Capstone Process Timeline. Students FAQs
Capstone projects are academic semester-long experiences for students nearing graduation. Student teams complete a substantial data science project that solidifies knowledge gained in the classroom and equips them for success in the following phase of their careers.
This fusion of classroom learning and work experience enables students to get ready for a career in industry and make the connection between theoretical ideas and practical applications.
Student teams make use of the technical and design skills they have developed throughout the Data Science curriculum to satisfy the sponsor’s objectives. They scope the issue and choose the most appropriate software, ML, and DL models to employ for it using standard software engineering and data science approaches. Then they create, evaluate, implement, and record their solution.
In order to assure project success and quality as well as to familiarize themselves with the process of professional software design, students also employ professional project management techniques.
All projects are carried out within the parameters of the course by student teams, under the direction of faculty, and in close coordination with the sponsor.
Corporations, small businesses, national laboratories,
R&D organizations, non-profit organizations and
faculty and staff members of the University of Colorado may become project sponsors.
|
Capstone projects courses are offered to final-year master's students.
|
Capstone sponsorship allows an organization to form an in-depth connection with a group of students who are nearing graduation. Through this connection, you may be able to find potential candidates and introduce them to the objectives and culture of your business. Capstone sponsorship serves as a mentorship role, helps our students become professionally ready, and instills positive habits as they get ready for the workforce.
In addition to your team, all students in the capstone course will be exposed to your organization as teams report out their work to the class. This offers excellent visibility, as capstone courses are large, and spreads word of mouth as students talk about their experiences throughout the year and at the university expo.
All project concepts should have a clear purpose with a recognized value to industry or society. They should have specific functional objectives and provide significant design challenges.
Projects must have a level of complexity that is compatible with a five to seven -person team of master's students working on average 10 hours each for 15 weeks in the spring semester. Students should be able to explore various design solutions and make choices based on sound engineering reasoning with creative latitude in arriving at a final design and implementation.
Exploratory or proof-of-concept projects can be quite successful as capstone projects.
Customer-critical initiatives cannot be approved as capstone projects unless the customer assumes full accountability for the result. Results that are "good-to-have" and "test-of-concept" studies are more likely to be appropriate. Although the goal is to make every project a success, the primary purpose of the capstone is educational. The University of Colorado cannot take any responsibility for results deemed by the customer as “insufficient.”
Sample Projects
Market basket analysis is the method to find the associations between the items / itemsets and based on those associations we can analyze the consumer behavior. We compared apriori, fp- growth, ECLAT and try to improve the fastest algorithm. We worked on transactional data of instacart and compared the algorithms which will future studies to choose algorithms based on their datasets. We named the new algoritms as fast apriori and super fast apriori.
If you are performing a workout and you need to evaluate your form, but do not have access or resources for an instructor to evaluate your form, it would be great to have an Artificial Intelligent agent provide real time feedback through your laptop or phone. We used a particularly accurate convolutional neural network called a faster-Region based CNN to find objects in a given frame of a video. We feed the object detection results of each frame into a recurrent neural network to track the objects throughout the video.
Predicting the likelihood of a customer abandoning an online shopping session without making a purchase can help marketers make quick marketing decisions. For example, if it appears that a user is unlikely to complete a transaction, they may be sent a tailored special offer based on their session history to entice them to do so. We use data from online shops about client shopping experiences to train deep learning models to predict shopping cart transaction.
Over 45,000 species of mushrooms have been identified in nature but only about 2,000 are edible. Due to their wide abundance, mushroom poisoning accounts for close to 70% of all natural poisoning and often leads to death. Our aim is using machine learning methods to characterize mushrooms with improved performance and that is robust to adding new features or species of mushrooms and explore more state-of-the-art neural network models for classifying mushrooms using images.
All sponsors are required to actively participate in the project they have funded. The project's sponsors should choose a Technical Lead from company who can commit to the project for at least one hour each week.
Close contact with the team during the early project definition phase is critical for project success. Depending on the software process model being utilized, the frequency of sponsor-team encounters will vary, and these are arranged jointly by the sponsor and team.
Course instructors are to be CC’d on all team contact.
Sponsors can choose to allow students to retain the IP from their work, or to retain all IP generated for the project,
Unless a project agreement is created based on a contract managed by the University’s Office of Contracts and Grants, all Intellectual Property (IP) rights resulting from the supported senior design project remain with the inventor(s), i.e. the students. All materials, software packages, etc. purchased to support the project will remain the property of the DS Department for possible future use in another project or class.
Participation in the course requires a financial commitment from most sponsors. University of Colorado Community non-profit organizations, and small businesses may apply to the Director of Senior Projects for a donation reduction or exemption.
Option 1: A $3,500 philanthropic donation made payable to the University of Colorado Foundation, to provide support to the University of Colorado Boulder Department of Data Science Senior Projects. This donation supports the Senior Projects class infrastructure and associated costs (instruction costs , software, computer labs, materials, supplies, disposables, posters, etc.)
Option 2: A $7,000 fee is charged if your organization wishes to retain project related IP. In this case a contract will be created through the University of Colorado Office of Contracts and Grants. Students assigned to these projects will be aware of the requirement to sign over all intellectual property rights to the sponsor.
With the help of course instructors, sponsors scope a project appropriate for master’s students and identify a technical lead who can interact with the team. In early January, projects are shared with students, who select their preferences. Instruction staff match students with projects according to preferences and skillsets required by the project.
In the first three weeks, approximately half of students’ time will be spent on coursework where they study requirements elicitation and analysis, software process models, systems engineering, software configuration management, risk management, team work, software documentation, IP law, and ethics. The remainder of the students’ time is focused on scoping and architecting a design approach to their team project. By the end of the first three weeks, students and sponsors will come to a written agreement as to the scope of the project and requirements for successful project completion.
Teams continue their work through the rest semester building, testing, and iterating on their models. At the end of the spring semester, students will present at our College of Engineering Expo, attended by thousands of students, faculty, and sponsors.
Teams give 3 presentations to the class in the semester and are subject to three reviews by the Project Review Board. Sponsors are welcome to attend any, or all, of these meetings.
All sponsors are required to actively participate in the project they have funded. The project's sponsors should choose a Technical Lead from company who can commit to the project for at least one hour each week. Close contact with the team during the early project definition phase is critical for project success. Depending on the software process model being utilized, the frequency of sponsor-team encounters will vary, and these are arranged jointly by the sponsor and team.
Course instructors are to be CC’d on all team contact.
Meet with the Capstone team as needed to understand sponsorship expectations, discuss project scope, and receive proposal paperwork. Once the teams have been formed, the student team will meet with the corresponding project sponsor in order to gain a deeper understanding of the project, sponsor goals, and confirm that the project, sponsor, and team are a good match.
The first task for all teams is to refine their understanding of the project, and the goals of their sponsor, to perform an initial risk evaluation, and identify the best data sciecne process model to use as a frame for developing the model. With these in place, teams will proceed to identify tools and technologies appropriate for the project and work with the sponsor to identify materials that constitute a complete project as appropriate for that specific project and according to the process model being used.
Teams design their solution, assign individual roles, and plan their project milestones for the remainder of the cycle. With the guidance of the sponsor, teams present their project design and may begin building. Sponsors are asked to submit feedback to instructional staff that contributes to student grades
Teams work throughout the semester to build the design that was prepared during project idea submission. Pivots or redesigns may occur with the guidance and permission of the sponsor technical lead. Students document their work, test for efficacy, and make recommendations for further work. Teams present their completed work, timeline for remaining work to the Project Review Board in the form of a presentation. Sponsors are strongly encouraged to attend the checkpoint presentations.
Teams present their work to the Project Review Board in the form of a presentation. Sponsors are strongly encouraged to attend the presentations. Sponsors are required to complete a team evaluation at the end of the spring semester. The end of semester evaluation forms will be used as the basis for the students’ course grades, and will be adjusted by the instructors according to peer evaluations and instructor observations to produce individual project grades.
Course is offered in both the fall and spring semesters, and second-year MS-DS on-campus graduate students are eligible to take the course.
Only 3 credits will count towards a degree.
Tution fee for course is same as any other 3-credit course. Please check the CU Boulder Bursar's Office for current tuition rates.
The course is almost ALL about the project. We only meet for presentations and trouble shooting. There are few lectures. No assignments. Just ONE project. The grading criteria is an integration of the feedback from instructor, peers, and the sponsors.
NO and NO. It is a course project. You can describe this experience as a “volunteer” project for a university course to avoid visa problems in the future.
All projects will be there for you to choose in week 1. You will list your preference of top 3, and the instructor will try to accommodate your preference as much as possible.
3 - 7 depends on the scope of the project. and it is average as 5.