Through the Looking Glass

Customer churn continues to be an issue of great importance for digital service providers, which harms revenues and long-term growth. The work presented here aims to utilize a machine learning point of view on user content interaction patterns over different digital platforms to predict churn. Temporally based engagement metrics, behavioral clustering and feature engineering is proposed to model user intent and detect users’ early signals of churn in an integration using the proposed framework. Supervised learning models are analyzed with Random Forest, Gradient Boosting, and Support Vector Machines among others on real world datasets that are available from streaming and SaaS platforms. It is demonstrated that the results help a lot in improving the prediction accuracy and early detection rates over traditional demographic and transactional models. Finally, this study provides a practical roadmap for digital service providers to adhere to data driven strategies and create counter measures that will increase user retention through the implementation of the proposed solution

The use of behavioral analytics and clustering to improve user retention tactics in user-centric systems is examined in this research. We create focused retention tactics that cater to certain user demands and behaviors by examining user behavior and grouping users into discrete groupings. According to our research, these data-driven tactics offer a scalable solution for various user-centric platforms while considerably increasing retention rates. Retaining customers has become critical for businesses hoping to prosper in today's fiercely competitive business environment. Businesses now need to concentrate on keeping their current clientele and acquiring new ones. Why? It is commonly known that boosting client retention rates can result in notable expansion and financial success for businesses. However, attaining high customer retention rates necessitates a calculated strategy and a thorough comprehension of your client's requirements and behavior.

In this report, we describe the experiences of students during the classroom implementation of a socially relevant project-based learning computer science (CS) course. We compared interviews from girls and boys in the course using thematic analysis to gain insight into ways to improve future course implementations and gain more interest from young women. Our findings include insights into how collaboration benefits girls, how reviewing introductory programming concepts is important to give all students the knowledge they need to succeed, and that socially relevant and interdisciplinary projects can increase all students interest in CS.

Existing approaches to teaching artificial intelligence and machine learning often focus on the use of pre-trained models or fine-tuning an existing black-box architecture. We believe advanced ML topics, such as optimization and adversarial examples, can be learned by early high school age students given appropriate support. Our approach focuses on enabling students to develop deep intuition about these complex concepts by first making them accessible to novices through interactive tools, pre-programmed games, and carefully designed programming activities. Then, students are able to engage with the concepts via meaningful, hands-on experiences that span the entire ML process from data collection to model optimization and inspection.

INFACT aims to provide differentiable teaching and learning activities in Computational Thinking (CT) that are inclusive for neurodiverse learners in grades 3-8. Neurodiversity refers to learners with autism, ADHD, dyslexia, and other related cognitive differences. Underlying many neurodiverse conditions is differences with executive function (EF). In previous research on game-based learning, neurodiverse students became class leaders in CT activities [1]. Special education teachers noted that CT practices align with strategies they use to support problem-solving practices. Leveraging these affordances of CT, INFACT activities are designed with embedded EF scaffolds to support neurodiverse learners.

There has been little research conducted about outcomes of children who have grown up with a parent with a disability. Much of the research that exists presupposes negative outcomes or is limited by small sample sizes that focus only on a specific parental disability. This eight year study was under

How do we integrate requisite thinking skills into children's earliest informal and formal learning experiences in order to prepare the next generation of problem solvers? This project examines how Computational Thinking (CT) skills align to the abilities and interests of young children and can be integrated with other STEM learning activities in order to promote school readiness. The goals of the project are to (1) build a knowledge base on how CT can be promoted in early childhood through integration with mathematics and science, (2) develop learning blueprints to guide the development of integrated classroom and family resources, and (3) develop and evaluate prototype activities to identify design principles useful to future development efforts integrating CT and STEM in early childhood. In this work, a multidisciplinary research team with expertise in CT, early science and mathematics, technology and media, and home and school learning partners with preschool educators and parents to co-design and pilot test prototype activities. Emerging prototypes presented integrate fundamental science and mathematics concepts and practices with CT skills such as problem decomposition, logical and algorithmic thinking, abstraction and debugging. They include hands-on activities with manipulatives familiar to young children and developmentally appropriate technology with unique affordances for complementing and reinforcing the learning that occurs during hands-on experiences. With a goal of addressing CS For All, the co-design, development and pilot testing of prototype activities aim to take into account the needs and strengths of preschool programs and families from low-income and culturally diverse communities.

To determine whether human immunodeficiency virus and acquired immune deficiency syndrome (HIV/AIDS), diabetes mellitus, and atopic disease are associated with herpes simplex virus (HSV) eye disease and to examine the characteristics of patients with HIV infection and HSV eye disease. Retrospective case-control study. A hospital-based control group was matched to outpatient visits from June 1, 2010, through May 31, 2014, at Montefiore Medical Center (Bronx, NY). Inclusion criteria were a diagnosis of HSV eye disease during the study period and residency in the Bronx. Associations evaluated included age, sex, HIV/AIDS, diabetes mellitus, and atopic disease. HSV eye disease was confirmed in 70 patients, who were compared with 280 controls. Patients with ocular HSV had a greater prevalence of HIV/AIDS compared with controls (8.6% and 2.9%, respectively). Using multivariate analysis to control for age and sex, atopic dermatitis [odds ratio (OR) 3.08, 95% confidence interval (CI), 0.84-11.20] and diabetes with chronic complications (OR 2.25, 95% CI, 0.91-5.61) approached significance, whereas HIV/AIDS (OR 3.37, 95% CI, 1.09-10.40), an age less than 45 years (OR 2.89, 95% CI, 1.54-5.41), and male sex (OR 1.85, 95% CI, 1.07-3.18) were significant. In patients with HIV infection and HSV eye disease, 3 of 6 (50%) had confirmed AIDS at the time of ocular HSV diagnosis. Patients with HIV infection have more than a 3-fold increased risk of developing ocular HSV.

Despite the ongoing debate about intervention best practices for intimate partner violence (IPV), few researchers have elicited the perspectives of clients themselves about what interventions most effectively decrease violence and increase safety. Using qualitative narrative analysis methodology, the researchers conducted 48 client participant interviews and 5 staff interviews to better understand couples' perspectives of a multicouple conjoint treatment program for IPV. Several recurring themes included (a) group purpose and general service characteristics, (b) motivation for participation, (c) comparison with other services, (d) benefits of, (e) disadvantages of, and (f) suggestions for Couples Achieving Relationship Enrichment. Important research implications for community intervention are discussed.