With massive availability of online learning resources on the Internet, a shift towards social and personalized learning at scale will transform the education industry for all.
CollegeShare is a social learning service that allows students and teachers across different colleges to share and discuss their learning, teaching, and college experiences around specific courses, while leveraging the power of analytics to achieve differentiated skills development, enhance collaborative learning, and inspire students to self-drive their learning experiences.
Context-Aware e-Collaborative Environments for e-Health Decision Support
This is so far the biggest ethnographic research project that I have led. The project involves field observation, interviews, diary study, and questionnaire survey of clinical practitioners across three geographical regions - the UK, the UAE, and Nigeria - in order to draw insights to inform the design e-Health systems for supporting collaborative clinical decision making among clinicians working independently across organizational and geographical boundaries.
The research uncovered that differences in clinical decision making practices are correlated with organizational cultures and geographical differences. But, it further indicates that they are associated with differences in local work contexts across work settings, and are moderated by adherence to best practice guidelines and the need for patient-centered care. The study further reveals that an awareness of three practice categories, namely the ontological, stereotypical, and situated practices plays a crucial role in adapting knowledge for cross-boundary decision support.
I designed and led the user research. I also designed and programmed the main solution architecture for the key project deliverable (the Context-Aware Decision support for e-HEALTH, CaDHEALTH system), which leveraged major technological innovations, such as the Web, the Grid, mobile networks, multi-dimensional image analysis, and intelligent systems for the creation of a pervasive, knowledge-driven, patient-centered, and easily accessible e-Health system for collaborative clinical decision support across organizational and geographical borders. I also proposed and demonstrated an approach that allows clinicians to semi-automatically evaluate the potential viability of a clinical practice behavior (e.g. a workaround) that may not adhere strictly to guidelines, but might be of potential benefit to a given clinical case, work context, or patient situation before applying the workaround in decision making
The project offered a set of design guidelines for the development of enterprise information systems for e-health decision support, and potentially for informing global e-Health strategies.
Understanding the Practice of Discovery in Enterprise Big Data Science
This project combined agent-based modeling with empirical field studies to understand and possibly predict how new data practices, organizational dynamics, and social infrastructures of existing platforms for big data science, shape innovation and scientific discovery in the era of big data. As we empower autonomous systems and algorithms driven by big data to make decisions on our behalf even in fuzzy areas that require the use of discretion and where there are no clearly defined right or wrong answers (e.g. to determine what to buy online, which movies to watch or airlines to fly, or who to befriend on Facebook), it becomes crucial for research to pay attention to the overarching goals, values, and practices that drive the functioning of our algorithms. For example, it is one thing for a machine learning expert or data scientist to say that their system is able to connect the dots and draw insights from data in order to determine that candidate X is not qualified for a loan, but it is a different thing to ask what belief (or value) systems govern the algorithm’s reasoning process? Toward what goal was the algorithm acting? Who generates the data? From where? Under what contexts? How do these ultimately affect the final outcome of the analytics process?
I initiated this project and led a team of three researchers to deliver on it. I designed the field study including observation and interview of data scientists aimed to understand how professionals (e.g., data analysts, medical researchers, etc.) discover and benefit from patterns in data. I developed a simulation model of big data science as a multi-objective function and an activity occurring within the social and organizational contexts of an enterprise.
Project results include the identification of new metrics for understanding collaborative discovery from data, and led to new insights for informing the design of tools, abstractions, and visualizations to facilitate collaborative discovery in big data science. The project also resulted in a paper presented at AHFE2015
Yes, you have now moved your business to the mobile platform. Your customers can access your services on the move via apps. Congratulations! But seriously, what's next?
To help you answer this question effectively is precisely what the goal of this project is. Your move to the mobile platform is a technological transformation. And along with that transformation, comes the cultural production of new forms of material practices. In order to derive maximum value from this transformation and remain competitive, there is a need to understand these practices through a user-centered study of your customers' behavior while using your mobile apps. Do they use the app as intended by your designers? Do they struggle with using the app? Where? Why? How does it affect your conversion rate?
We analyze large-scale data about app user behaviors (screen navigation patterns, gestures, session time, screen views, etc.) — collected using mobile analytics iOS and Android SDK tools — to derive value from usage data, deliver insights about your how the app is used, and make recommendations about improving your app design.
I developed a solution for automatically extracting and parsing the iOS SDK unstructured data about mobile user behaviors into structured dataset of input features for doing scalable machine learning and behavior analytics. I also designed the research study integrating both quantitative/statistical analysis of iOS SDK data and qualitative analysis of data obtained via user interviews
This project is still on-going. Preliminary result includes a patent disclosure, as well as successful initial user testing with clients.
The Work Exchange
I led the technical development of the Work Exchange — an intra-organizational crowdsourcing system that provides a virtual exchange where work is traded between those who need work done and those who have the capabilities to do the work across functional boundaries of (large) organizations.
Imagine -- for example, as a team manager -- the possibility of tapping into a virtual work exchange throughout your enterprise, or even across enterprise boundaries with people around the world who have unique skills in order to draw talents as and when needed to get your work, thereby beating the limitations of existing traditional staffing models. The Work Exchange is a social business platform that achieves that vision using market mechanisms. The system enables fast and scalable design of work requests by componentizing units of tasks as service requests, uses market-based matching mechanisms – producer selection, bids, auctions and contests, and adopts social identity for expertise recognition, social reputation management, and reward.
From a research point of view, the project explored a deeper understanding of human experience at work, and involved in-depth analyses of organizational work practices, user-centered research and design, service system modeling, and concept development.
The Work Exchange was deployed for pilot use at IBM's Global Organizational Change Management Unit with hugely successful results.