X-Lab Project Opportunities
Do you have data that needs to be analyzed? A machine learning model that needs to be tested? An app developed? BU Spark! offers a unique opportunity for organizations to partner with Boston University students with computer science and engineering skills on their technology projects. There are several no-cost or low cost options to engage students, see the various options listed below:
- Free Class Projects
- Paid On-Campus Externships
- User Design/User Interaction support
BU Spark! Free Class Projects
Each semester, BU Spark! places computer science and engineering students on data science, machine learning, and software development projects provided by external partner organizations. Projects are approximately 60 hours of work involving teams of three students and require external partner organizations to dedicate approximately 45 minutes per week for the 8 weeks of project implementation. You can Apply To The X-Lab HERE and find FAQs below.
XCC433: Justice Media Co-Lab Instructors: Brooke Williams and Osama Al-Shaykh
The Justice Media Co-Lab, a collaboration between BU Spark!, based at the Faculty of Computing & Data Sciences, and the Journalism Department at the BU College of Communication, matches interdisciplinary student teams with a background in computer science, statistics, computer engineering, or data science or journalism related disciplines with computational journalism projects provided by external media partners.
Projects vary in size and scope and range from smaller projects that are approximately 60 hours of work to larger projects that can be as much as 100 hours of work per team member over the course of a semester. Projects are comprised of teams of approximately 3 computer science or engineering students and require partner organizations to dedicate approximately 45 minutes per week for the 8-12 weeks of project implementation.
Data Science Class
Data science projects are designed to help partners answer strategic questions from data analysis conducted on large and multiple data sets. Projects must include data collection (e.g. calls to an API, parsing/ crawling web pages, etc.); data compilation and cleaning (e.g. combining with an internal spreadsheet), and analysis. Students will then analyze the data using methods such as clustering, classification, regression, and network analysis. Ideally, projects include multiple data sets comprised of a minimum of ~5,000 records. It is important that the sponsoring organization provide the specific question or questions they are seeking to have answered from the analysis of these data.
Data Science Project Example: In partnership with the ACLU-MA, Spark! students were able to identify patterns, including racial and geographic disparities based on the treatment residents received by the Boston Police Department. Students used three data sets to arrive at their conclusion including stop and frisk data, crime incident report data, and census data.
Data Science Project Example: Using census data and business license database, Spark! mapped residential concentrations and economic development activity among Brasilians in Boston for Digaai.
Machine Learning Class
Machine learning projects aim to build algorithmic models, based on supervised learning approach, that accurately forecast desired outcomes. Projects must include already compiled and cleaned data sets comprised of a minimum of ~5,000 records that includes a portion of “groundtruth” data, i.e. where the result is known. Students will use this groundtruth to build a model that trains the remaining data to achieve forecasts based on a desired threshold of accuracy determined by the partner organization.
Machine Learning Project Example: Working on behalf of an early-stage start up and using data pulled from the Zillow API and RMLS Data, students created an algorithm to estimate real estate sales price. Variables included interior and exterior photos, location, and house features. The model built by the students achieved an 81% accuracy rate.
Machine Learning Project Example: In order to help Converse understand the effectiveness of its product launches BU Spark! students built a model that helps to track social media traffic once a product is launched. Measurements included but are not limited to, demographics and geography of the traffic interested in the product and how the information about the product launch is shared over time.
CS 791: Applied ML For Public Health Instructor: Elaine Nsoesie
Machine learning methods are being used in the analysis of data (e.g., text, image, sound, video and biological data) to understand disease and health trends, and to improve individual and population health. The goal of this course is to provide students with prior knowledge of Machine Learning techniques with the opportunity to gain hands-on experience working on a data analysis project. Throughout the semester, students will work in groups to apply Machine Learning to solve a specific problem submitted by one of our research partners. Weekly lectures will focus on exposing students to (a) applications of ML to public health problems, (b) openly available computational resources, and (c) ongoing research by experts working in ML and health. The course is open to both Masters and PhD students.
Spark! X-Lab Practicum Class
Projects for this class are more open-ended and can include web or mobile app development, data science, data visualization, machine learning and more. The projects accepted will be based on the skills of the students taking the course.
Practicum Project Example: On behalf of an Ed Tech startup, BU Spark! created a prototype of an AR/VR app designed to stimulate children to utilize their immediate environment to explore key early math concepts in geometry.
Practicum Project Example: BU Spark! working in collaboration with a local investigative journalist, built an automated scraper to extract data from public websites into a customized database accompanied by a searchable user interface where reporters or citizens can sort, search and otherwise explore the data.
Paid On-Campus Externship Projects
In addition to the class projects, partners can choose to fund a team of students through an on-campus Spark! Externship. Like the X-Lab Practicum class above, these projects offer greater flexibility in terms of technical scope but they also offer greater flexibility in terms of timing because projects are accepted on a rolling basis.
Here’s how the X-Lab works: Please email firstname.lastname@example.org if you have any questions about the X-Lab. Applications are accepted on a rolling basis.
- Submit A Project Application HERE!
- BU Spark! will schedule a consultation call to discuss the best fulfillment option.
- Scope is refined and agreed upon.
- Spark! recruits a team of BU students or matches the project to a class.
- Student team works directly with the project partner with oversight from Spark!
- Project is delivered to the partner along with any relevant code, data, or other agreed-upon deliverables.
User Design/User Interaction support
In addition to the paid on campus externship, partners can choose to fund a student through an on-campus Spark! Externship, to work on wireframes for your project. The wireframes are often needed before a project can be developed in a technical environment e.g. mobile application, web development projects.
We have three different project options:
1) UX Prototype and Sketch Wireframe: these are rough sketches of your product concept that reflect your priority functionality for the purposes of testing with users and/or sharing with your development team
2) Style: you have wireframes, you know what you are building or are already building it, but you want to develop a brand identity and a style guide for your product. This includes a logo, color scheme, font selection, etc.
3) Hi-Fidelity Wireframe Designs: you know what you are building or are already building it, you have your style guide and you need help applying the style guide to your wireframes to integrate with your front end development
We offer greater flexibility in terms of timing because projects are accepted on a rolling basis.
Applications are accepted on a rolling basis.
|Q: What types of projects does BU Spark! accept?|
|A: We are interested in all projects from all sectors. Last semester we received 47 project applications from 40 different organizations, including non-profits, faculty research, startups, local government, and global companies. Spark! is an experiential learning opportunity for students who are still developing their skills. We prefer exploratory projects that are not mission critical to your organization.|
|Q: What are the costs?|
|A: There are a variety of options for getting your projects completed. Your project may be completed by a student team who will work on your project as part of a required class assignment, at no cost. These projects tend to be focused on a few specific technical areas and must be completed during the specified time.
Alternatively, you may elect to have your project completed as a student consulting project, on a fee for service basis. These projects have greater flexibility regarding scope and timing as well as the range of technical expertise required. The primary requirements are funding and designation of a point person from the organization who can oversee the work of the student team.Finally, we are seeking App development projects from nonprofits that must be submitted by September 1st. These projects do not require compensation.
|Q: Can you guarantee that my project will be matched to a team of students?|
|A: No, we can’t guarantee that projects will be matched. We will do our best to help you scope projects that fit within the requirements of the chosen track, fall within the skillset of our student community, and present a compelling opportunity to students. We will also publicize with our student community, but we are unable to make any guarantees that the projects will be adopted. Incentives and access to mentors always help!|
|Q: What if I’m unable to answer some of the questions on the project proposal form?|
|A: No problem, indicate in the application where you have questions and we’ll contact you to clarify any missing information. If you need assistance or have any questions, you can email us at email@example.com..|
|Q: What are the requirements for serving as a project mentor?|
|A: At minimum, we need a point person in the partner organization with the context necessary to provide direction to student on the expected project outcomes. The mentor does not necessarily need to be technical, but it is important that this individual can provide a reasonable amount of time to guide the students through the project.|
|Q: Who Owns the IP? What about NDAs? Will BU Spark! provide a legal agreement?|
|A: We encourage our partners to allow the students to retain ownership or use of the methodologies they develop for your project under an open source licensing agreement. However, we understand that you may want ownership of the specific application or analysis completed on your behalf to remain proprietary or non-disclosed. We also understand you may want to own the IP/ methodologies developed by the students. In short, we encourage your organization to communicate your expected terms for this partnership in advance to ensure that all agreements are within BU’s policies and procedures.
Spark! has template agreements that can be used and adapted based on the required terms. BU Spark! cannot sign agreements on behalf of students, so this agreement will be between your organization and the students, with no legal liability to BU Spark!
|Q: Can I get the code the students develop?|
|A: Yes, we have created a Github repository where the code from all student projects will be uploaded. We can give you access to this repository. Alternatively, if there is sensitive information you would not like to be uploaded to Github, students can send you this part of their code separately.|
|Q: Will BU Spark! provide a legal agreement?|
|A: Yes, we have template agreements that can be used and adapted based on the required terms. BU Spark! cannot sign agreements on behalf of students, so this agreement will be between your organization and the students, with no legal liability to BU Spark!|
|Q: What happens if the student team doesn’t finish my project?|
|A: We hope you will always be satisfied with the outputs from the partnership. However, there is always a risk that the volume of work may exceed the capacity of the students from either a time or technical skill perspective. The best way to mitigate against this risk is to work in close collaboration with BU Spark! and your student team to accurately scope the project before you get started. You will need to communicate regularly with students your top priorities for the project.|
|Q: What happens if the quality of the work by the student team does not meet expectations?|
|A: We will try our best to make sure this does not happen and we do our best to make sure students are vetted before they are matched to the client project. There will be several mid-project assessments and deadlines during the projects as well in which will allow you to correct for quality. If you are having challenges with the student team – either in terms of regular communications or work quality – please let BU Spark! know. This is an experiential learning opportunity for students and you will have an opportunity to complete a final evaluation so BU Spark! and the students can learn from the engagement.|
Spark! X-Lab FAQs – Class Projects
|Q: What are the requirements for class projects?|
|A: There are several requirements for class projects, including:
1. At the beginning of the semester, we ask that someone from your organization participate in an informal presentation to students, during which you will give a 10-minute overview of your organization, your proposed project, and answer any questions students may have about the project.
2. Provide any key data or documentation necessary for students to facilitate completion of their project.
3. Provide a mentor(s): You or someone from your organization will serve as an advisor to the student team(s) assigned to your project to make sure they are on track and meeting your brief. Ideally, the mentor would complete weekly check-ins with the student teams either in-person or virtually during the eight week period when students are actively working on the project.
4. If possible, participate in an end-of-semester poster presentation event: Students present their finished projects to the class and partners.
5. Complete an evaluation form for the student team and provide feedback on the Spark! Partner Projects program.
|Q: What types of projects can be completed as part of the class projects?|
|A: The class projects vary every semester depending and have specific requirements. You may visit our website for more information. Examples of projects done in the past are:
· Data collection and analytics on the popularity of two types of Converse sneakers based on Twitter
· Analysis of Boston Police Department’s crime and policing data for the ACLU of Massachusetts to examine the fairness of policing behaviors across different regions.
|Q: What do I need to prepare for the presentation of my project to students?|
|A: Expect 5 minutes to describe your organization and project, and 5 minutes for Q&A from the class.|
|Q: What if I am unable to attend the presentations to students or am not based in Boston?|
|A: If you can’t be there in person, you can present via Skype or simply record a video of your presentation. The initial presentations are not required, but will greatly enhance your chances of getting your project matched.|
|Q: Do I need to prepare anything for the Poster Presentation Session? What if I won’t be able to attend the Poster/ Final Presentation Session?|
|A: No, you do not. If you wish, you may have the students present their final work to you at a different time. This will need to be scheduled by mid-semester. Students will also submit a final report to accompany the end-of-semester presentation.|