Author Archives: Evelina Rosser

HOW ARE CAPSTONE PROJECTS EVALUATED AT UCF

Leave a reply

Capstone projects at UCF are meant to demonstrate a student’s mastery of the key concepts, skills, and knowledge learned throughout their undergraduate academic program. With that goal in mind, capstone projects undergo a rigorous evaluation process to ensure students are assessed in a comprehensive manner.

At the start of the capstone experience, students work closely with their capstone instructor and other faculty advisors to determine an appropriate project topic that aligns with their major and allows them to apply what they have studied. Topics can range widely depending on the discipline but all must be substantive enough to require integrating learning from multiple courses and demonstrating advanced skill levels. The topic selection is initially reviewed and approved by capstone instructors.

Once a topic is chosen, students develop a detailed project proposal outlining the goals, scope, methodology, timeline, and anticipated outcomes or deliverables of their planned work. Proposals are typically 5-10 pages and include elements such as an introduction and problem statement, literature review, proposed methods, intended results or product, and overview of how the project will be evaluated. These initial proposals are critically reviewed by capstone instructors and often other relevant experts. Feedback is provided to ensure the proposed work is properly focused, sufficiently ambitious in its goals yet realistic in its approach. Students may need to revise and resubmit proposals until receiving full approval to move forward.

With an approved proposal in hand, students then embark on executing the key aspects of their capstone project work over one or two semesters. Throughout this period, students meet regularly with their capstone instructor and other advisors for guidance, mentorship, and to track progress. Capstone faculty review draft deliverables, provide substantive feedback for improvement, and hold students accountable to their proposed timeline and standards of quality. Midway through, students often submit an interim report on accomplishments and any adjustments needed to their original proposal.

Toward the end of the capstone term, students submit a final comprehensive written report, portfolio, thesis, or other culminating product, adhering to prescribed formatting guidelines. The quality and rigor of these final deliverables are of paramount importance, as they serve as the primary basis for evaluation. Accompanying materials such as annotated bibliographies, datasets, code, prototype designs, marketing or outreach plans, etc. provide further evidence of the work and often factor into final grades.

Final capstone projects also typically include a public presentation or defense. This allows students to orally communicate about their work to a broader audience, including capstone instructors, other faculty members, student peers, and often community stakeholders or employers. Presentations are usually 15-25 minutes followed by a lengthy question and answer session where presenters must demonstrate expertise in both their project substance and ability to think on their feet.

The capstone evaluation process at UCF is intended to comprehensively judge student performance across multiple critera, including but not limited to:

Depth and quality of research, analysis, or other technical work conducted
Clear identification and importance of the research question/problem addressed
Appropriate selection and application of relevant conceptual frameworks/theories
Thoroughness and effectiveness of proposed and implemented methodsologies
Rigor of data collection, measurement, analysis techniques as applicable
Strength and validity of results, insights, conclusions reached
Clarity, organization, and quality of writing in the final report/deliverables
Effectiveness of oral presentation skills as demonstrated in defenses
Ability to handle questions that may challenge conclusions or point out limitations
Extent to which the work makes an important contribution to the relevant field
Demonstration of initiative, independence, and advanced skill mastery
Adherence to deadlines, formatting requirements, and other expectations

Capstone instructors and reviewing faculty utilize detailed rubrics to systematically evaluate student performance across these criteria when determining final grades. Rubrics include quantitative scoring of elements as well as opportunities for qualitative commentary. Scores on deliverables, presentations, and other factors such as peer/self evaluations are combined mathematically according to predetermined weightings. Students must meet minimum thresholds across criteria in order to pass. Those whose performance far exceeds expectations can earn A grades, while substandard work may result in no course credit.

The capstone evaluation process at UCF aims to provide a comprehensive, transparent and rigorous assessment of student achievement through significant applied works of independent scholarship. By design, the capstone experience cultivates advanced research, technical and soft skills while confirming whether undergraduates have gained the knowledge and abilities befitting degree conferral. The multi-stage process of proposal development, ongoing guidance, and summative evaluation through rubrics helps ensure this important learning outcome is realized for all students.

HOW CAN INDIVIDUALS CONTRIBUTE TO REDUCING GREENHOUSE GAS EMISSIONS

Leave a reply

Transportation is a major source of greenhouse gas emissions for many people. Individuals can choose more sustainable transportation options to help lower their carbon footprint. Walking, biking, carpooling or taking public transit when possible are excellent low-carbon alternatives to driving alone. For longer commutes when other options aren’t feasible, driving a fuel-efficient vehicle, such as a hybrid, can help reduce emissions. Maintaining proper tire pressure and driving habits like avoiding excessive idling also improves gas mileage. Some people may be able to reduce personal vehicle use, through teleworking if their job allows it, living closer to amenities or dedicating a few days a week to avoiding car trips. For those who must drive, electric vehicles are becoming more mainstream and practical for many lifestyles, providing a zero-emissions way to drive.

When it’s time for a new vehicle purchase, choosing one with the highest fuel economy or that runs on alternative fuels or electricity will lock in emissions reductions for years of use compared to continuous driving of a gas-guzzling vehicle. Additionally, individuals can support policies that encourage the development of electric vehicles and alternative fuels, as well as expand public transit and active transportation infrastructure to offer more low-carbon options. Writing to elected representatives about climate-friendly transportation priorities is one way to create policy change.

At home, energy use for heating, cooling, appliances and other household needs accounts for a large portion of residential emissions. Implementing energy efficiency measures is one of the fastest and most affordable ways for individuals to cut carbon. Simple steps include weatherizing homes by adding insulation and sealing air leaks, installing programmable thermostats andLED lighting, and utilizing smart power strips. Transitioning home appliances to the most efficient models available during replacement cycles and air drying clothes instead of running lengthy dryer cycles also shaves emissions. Individual choices about home size and location can factor into emissions too – multi-family housing and smaller homes typically have lower energy needs than larger single-family units. Living in a more compact, walkable community near amenities and work reduces transportation demands.

For homeowners, investing in renewable energy sources like solar panels can allow a transition away from fossil fuel-derived electricity over time. Renting property may limit direct investment options, but renters still have opportunities through energy efficiency actions and choices about where to live. Supporting utility or statewide clean energy policies and programs through advocacy or by opting into green energy rate structures can also help scale up renewable infrastructure that benefits all customers. At the federal, state and local level, lobbying representatives to strengthen building codes and energy standards boosts broader emissions progress.

Dietary choices represent another major lever individuals have for lowering their carbon footprint. Producing and transporting meat, especially beef, generates more greenhouse gas emissions than producing plant-based proteins like beans, lentils and vegetables. Shifting toward a diet rich in whole grains, fruits and vegetables while moderateing or eliminating red meat if possible can significantly curb an individual’s food-related emissions. When eating meat, prioritizing chicken, fish and eggs over beef provides an easier reduction. Reducing food waste by mindful shopping also prevents emissions from uneaten food going to landfills.

In terms of consumer purchases overall, individuals have the option to favor durable, high-quality, locally-made goods that can be repaired rather than frequently replaced. This helps avoid high upfront and continual embedded emissions from manufacturing, shipping and discarding products with short lifespans. Staying up to date on sustainability product reviews enables choosing appliances, electronics and other items with efficient or recycled materials. When old items must retire, donating or recycling them diverts material waste from landfills. Minimizing consumption and single-use plastics also lightens environmental impacts. On a broader scale, civic engagement and voting for representatives prioritizing climate solutions influences policy and infrastructure support for a greener economy.

The daily and long-term choices outlined here demonstrate that individuals have powerful collective ability to shape systems and drive demand in a lower-carbon direction when acting on options available through lifestyle, advocacy and consumer power. While societal changes also depend heavily on coordinated climate policy and actions across governments and industries, individual actions can make meaningful contributions to emissions reductions when started early and sustained over lifetimes. With creative problem-solving approaches tailored for different circumstances, opportunities exist for people everywhere to participate in climate solutions through daily living. While no one action alone solves climate change, the combined efforts of conscientious individuals transitioning toward lower-impact choices represent important momentum for building a sustainable future together with broader policy support.

HOW DID YOU DETERMINE THE FEATURES AND ALGORITHMS FOR THE CUSTOMER CHURN PREDICTION MODEL

Leave a reply

The first step in developing an accurate customer churn prediction model is determining the relevant features or predictors that influence whether a customer will churn or not. To do this, I would gather as much customer data as possible from the company’s CRM, billing, marketing and support systems. Some of the most common and predictive features used in churn models include:

Demographic features like customer age, gender, location, income level, family status etc. These provide insights into a customer’s lifecycle stage and needs. Older customers or families with children tend to churn less.

Tenure or length of time as a customer. Customers who have been with the company longer are less likely to churn since switching costs increase over time.

Recency, frequency and monetary value of past transactions or interactions. Less engaged customers who purchase or interact infrequently are at higher risk. Total lifetime spend is also indicative of future churn.

Subscription/plan details like contract length, plan or package type, bundled services, price paid etc. More customized or expensive plans see lower churn. Expiring contracts represent a key risk period.

Payment or billing details like payment method, outstanding balances, late/missed payments, disputes etc. Non-autopaying customers or those with payment issues face higher churn risk.

Cancellation or cancellation request details if available. Notes on the reason for cancellation help identify root causes of churn that need addressing.

Support/complaint history like number of support contacts, issues raised, response time/resolution details. Frustrating support experiences increase the likelihood of churn.

Engagement or digital behavior metrics from website, app, email, chat, call etc. Less engaged touchpoints correlate to higher churn risk.

Marketing or promotional exposure history to identify the impact of different campaigns, offers, partnerships. Lack of touchpoints raises churn risk.

External factors like regional economic conditions, competitive intensity, market maturity that indirectly affect customer retention.

Once all relevant data is gathered from these varied sources, it needs cleansing, merging and transformation into a usable format for modeling. Variables indicating high multicollinearity may need feature selection or dimension reduction techniques. The final churn prediction feature set would then be compiled to train machine learning algorithms.

Some of the most widely used algorithms for customer churn prediction include logistic regression, decision trees, random forests, gradient boosted machines, neural networks and support vector machines. Each has its advantages depending on factors like data size, interpretability needs, computing power availability etc.

I would start by building basic logistic regression and decision tree models as baseline approaches to get a sense of variable importance and model performance. More advanced ensemble techniques like random forests and gradient boosted trees usually perform best by leveraging multiple decision trees to correct each other’s errors. Deep neural networks may overfit on smaller datasets and lack interpretability.

After model building, the next step would be evaluating model performance on a holdout validation dataset using metrics like AUC (Area Under the ROC Curve), lift curves, classification rates etc. AUC is widely preferred as it accounts for class imbalance. Precision-recall curves provide insights for different churn risk thresholds.

Hyperparameter tuning through gridsearch or Bayesian optimization further improves model fit by tweaking parameters like number of trees/leaves, learning rate, regularization etc. Techniques like stratified sampling, up/down-sampling or SMOTE also help address class imbalance issues inherent to churn prediction.

The final production-ready model would then be deployed through a web service API or dashboard to generate monthly churn risk scores for all customers. Follow-up targeted campaigns can then focus on high-risk customers to retain them through engagement, discounts or service improvements. Regular re-training on new incoming data also ensures the model keeps adapting to changing customer behaviors over time.

Periodic evaluation against actual future churn outcomes helps gauge model decay and identify new predictive features to include. A continuous closed feedback loop between modeling, campaigns and business operations is thus essential for ongoing churn management using robust, self-learning predictive models. Proper explanation of model outputs also maintains transparency and compliance.

Gathering diverse multi-channel customer data, handling class imbalance issues, leveraging the strengths of different powerful machine learning algorithms, continuous improvement through evaluation and re-training – all work together to develop highly accurate, actionable and sustainable customer churn prediction systems through this comprehensive approach. Please let me know if any part of the process needs further clarification or expansion.

HOW OFTEN SHOULD THE STRATEGIC PLAN BE REVIEWED AND UPDATED TO ENSURE ITS EFFECTIVENESS

Leave a reply

Strategic plans are designed to help organizations achieve long-term goals and objectives, but for a strategic plan to remain relevant and guide an organization effectively, it needs to be reviewed on a regular basis and updated when necessary. The optimal frequency for reviewing and updating a strategic plan can vary depending on factors like the organization’s industry, size, resources, and rate of change in its external environment. Most experts recommend conducting comprehensive reviews of the strategic plan at least once a year, with some interim reviews throughout the year as well.

Conducting an annual review allows an organization to assess progress made against the strategic plan on a regular cadence. It provides an opportunity to revisit goals, objectives, strategies, and initiatives outlined in the plan and evaluate whether they are still appropriate given changes that may have occurred internally or externally over the past year. An annual review meeting typically involves gathering key stakeholders from across the organization who were involved in developing the original plan. During the meeting, participants discuss what strategic priorities and tactics worked well over the past 12 months and which may need refining. They also look at whether the overall vision and mission still align with the organization’s current direction or if updates are warranted. Data on key performance indicators is analyzed to determine what strategic priorities drove the most success and where improvements are needed. The annual review culminates with an assessment of whether any elements of the plan, such as timelines, budgets, or departmental responsibilities need modification to optimize results over the coming year.

While an annual comprehensive review provides the necessary periodic check-in, some organizations also find value in conducting interim reviews on a quarterly or biannual basis. These shorter check-ins allow for more frequent monitoring of progress against objectives and timelines outlined in the plan. They provide opportunities to course correct sooner if implementation is lagging or external factors arise requiring an adjustment of strategic priorities mid-year. During interim reviews, participants typically focus the discussion on a subset of strategic initiatives, priorities or key performance indicators to keep the meetings efficient. Any recommended changes uncovered during an interim review would then be documented and fully evaluated during the next annual review meeting when a comprehensive refresh is conducted if needed.

For organizations operating in dynamic industries or markets that change rapidly, it may even make sense to review the strategic plan on a semi-annual basis to ensure it remains optimally aligned. Reviews that are conducted too frequently, such as monthly, run the risk of disrupting implementation efforts by constantly refining priorities before they have had enough time to take hold. There also needs to be a balance between reviewing frequently enough to stay nimble without expending too many resources on the review process itself.

The timing of annual reviews is also an important consideration. Most experts recommend scheduling the annual strategic plan review meeting towards the end of the fiscal or calendar year, typically in the last quarter. This allows time following the meeting to refine implementation plans for the coming year based on insights from the review. It also provides a natural checkpoint at the close of the year to evaluate performance and progress made against the existing plan. Some organizations find value in conducting a portion of the annual review mid-year as well to incorporate any learnings or adjustments into the second half implementation.

Regardless of review frequency or timing, it is critical that strategic plan reviews involve gathering input from leaders and contributors across all divisions and levels of the organization. Getting diverse perspectives is important for identifying opportunities or risks that may not be as obvious from an executive level view. The review process also needs to incorporate analysis of both qualitative and quantitative performance data to ensure any recommended updates to strategies or priorities are firmly grounded in facts rather objective opinions. With regular, systematic reviews built into the process, an organization’s strategic plan has the best chance of remaining an effective roadmap to drive long-term success even as internal or external conditions inevitably change over time.

Most experts agree that reviewing a strategic plan at minimum on an annual basis, with some organizations benefitting from additional interim reviews quarterly or biannually, provides the necessary cadence to evaluate progress and ensure the plan remains optimally aligned. The overriding goal of maintaining a regular review schedule is to continuously refine implementation strategies based on learnings so the organization can dynamically respond to opportunities while navigating challenges to stay on track with its long-term vision.

CAN YOU PROVIDE SOME EXAMPLES OF MACHINE LEARNING CAPSTONE PROJECTS THAT STUDENTS HAVE WORKED ON

Leave a reply

NLP sentiment analysis of restaurant reviews: In this project, a student analyzed a dataset of thousands of restaurant reviews to determine the sentiment (positive or negative) expressed in each review. They trained an NLP model like BERT to classify each review as expressing positive or negative sentiment based on the words used. This type of sentiment analysis has applications in determining customer satisfaction.

Predicting bike rentals using weather and calendar data: For this project, a student used historical bike rental data along with associated weather and calendar features (holidays, day of week, etc.) to build and evaluate several regression models for predicting the number of bike rentals on a given day. Features like temperature, precipitation and whether it was a weekday significantly improved the models’ ability to forecast demand. The models could help bike rental companies plan fleet sizes.

Predicting credit card fraud: Using a dataset of credit card transactions labeled as fraudulent or legitimate, a student developed and optimized machine learning classifiers like random forests and neural networks to identify transactions that have a high likelihood of being credit card fraud. Features included transaction amounts, locations, and other attributes. Financial institutions could deploy similar models to automatically flag potentially fraudulent transactions in real-time.

Predicting student performance: A student collected datasets containing student demographics, test scores, course grades and other academic performance indicators. Several classification and regression techniques were trained and evaluated on their ability to predict a student’s final grade in a course based on these factors. Factors like standardized test scores, number of absences and previous GPA significantly improved predictions. Such models could help identify students who may need additional support.

Diagnosing pneumonia from chest X-rays: In this project, a student analyzed a large dataset of chest X-ray images that were manually labeled by radiologists as either having signs of pneumonia or being healthy. Using techniques like convolutional neural networks, they developed models that could automatically analyze new chest X-rays and classify them as showing pneumonia or being normal with a high degree of accuracy. This type of diagnostic application using deep learning has real potential to help clinicians.

Predicting housing prices: A student collected data on properties sold in a city including features like number of bedrooms, bathrooms, lot size, age and neighborhood. They developed and compared regression models trained on this data to predict future housing sale prices based on property attributes. Factors like number of bathrooms and lot size significantly impacted prices. Real estate agents could use similar models to estimate prices when listing new homes.

Recommending movies on Netflix: Using Netflix’s anonymized movie rating dataset, a student built collaborative filtering models to predict rating scores for movies that a user has not yet seen based on their ratings history and the ratings from similar users. Evaluation metrics showed the models could reasonably recommend new movies a user might enjoy based on their past preferences and preferences of users with similar tastes. This type of recommendation system is at the core of how Netflix and other platforms suggest new content.

Predicting flight delays: For their project, a student assembled datasets containing flight records along with associated details like weather at origin/destination airports, aircraft type and airline. Several classification algorithms were developed and evaluated on their ability to predict whether a flight will be delayed based on these features. Factors like temperature inversions, crosswinds and aircraft type significantly impacted delays. Airlines could potentially use such models operationally to plan for and mitigate delays.

Predicting diabetes: Using medical datasets containing biometric/exam results of patients together with diagnoses of whether they had diabetes or not, a student developed and optimized machine learning classification models to identify undiagnosed diabetes cases based on these risk factor features. Features with the highest predictive value included BMI, glucose levels, blood pressure and family history of diabetes. Physicians could potentially deploy or consider similar models to help screen patients and supplement their clinical decision making.

As demonstrated through these examples, machine learning capstone projects provide students opportunities to work on real-world applications of their skills and knowledge. Some key benefits of these types of projects include: gaining hands-on experience applying machine learning techniques to solve problems, developing skill in data preparation, feature engineering, model development/evaluation and interpretation. They also help students demonstrate their abilities to potential employers or for further academic studies. Capstone projects are an ideal way for students to showcase what they’ve learned while working on meaningful problems.