S-Graffito project is aimed at building a streaming graph management system that addresses the processing of OLTP and OLAP queries on high streaming rate, very large graphs. These graphs are increasingly being deployed to capture relationships between entities (e.g., customers and catalog items in an online retail environment) both for transactional processing and for analytics (e.g., recommendation systems). The project consists of two components: streaming graph querying and streaming graph analytics. I have been involved in the second component (streaming graph analytics) during my PhD and PostDoctoral fellowship at the University of Waterloo.

Towards RELIABLE S.t..r...e....ami..ng Graph Analytic

Continuing my PhD studies (published in sGrow and sGrapp papers), I am developing simultaneously effective, efficient, and privacy-preserving solutions for streaming graph drift detection (sGradd) and streaming graph drift prediction (sGradp). The problem is concept drift (CD) detection: CD is a phenomenon that occurs when ``changes in hidden context can induce more or less radical changes in target concept''. Hidden context refers to insufficient, incomplete, or unobservable information about input data. Target concept refers to known and/or observable information that have direct impact on the task's output. Our goal is signaling the changes in the hidden contexts such as the generative source(s) of streaming (graph) data records.

Prompt CD management in streaming settings is important for generating relevant, reliable, and effective outputs. CD detection and understanding, benefit (i) development of accurate generative data models which are expected to (not) preserve concepts, (ii) generalization of analytics, (iii) designing algorithms (e.g., network protocols for error control and estimating the time-to-live of routing packets), and (iv) anomaly detections in areas such as the following.

data management systems: erroneous data for buffer management in streaming graph management systems
security: data manipulation in communication links
health care and bioinformatics: irregular data for emergency health conditions
e-commerce and web systems: unexpected data for supply chain management
distributed systems, social networks, computational networks, network on chips, cellular backhaul and core networks: artificial and abrupt traffic data for network traffic control
environments, geological and geospatial analytics: abnormal data for map routing technologies, analysis of mineral by-products, geohazard assessments
multimedia software development: unprecedented data for musical contests and athletic performance evaluation (e.g., in artistic gymnastic and taekwondo matches), sound control, speaker view, and bandwidth adjustment in videoconferencing tools
neuroscience: neuronal spikes for cognitive neuroscience studies

a Data System for Coordinating Seminars (DSCS)

a DS for Seminar Coordination

Please take a look at these slides to learn more and leave comments therein.

Running a seminar series such as DSG Lunch Talks, where researchers from different disciplines and with different schedules get together, listen to a research talk, and share research ideas, requires coordinating multiple items. Examples include schedule planning, booking meeting room(s), (re)booking seminars, communicating seminar information, ordering/delivering/serving food, seminar announcements, onboarding participants, question answering, expense management, time management, advertisement, and seminar streaming. Coordinating these tasks manually demands hours and even days of work and effort and working with several people for each seminar.

According to my ~four years of experience coordinating the aforementioned tasks for a wide variety of weekly seminars in the UWaterloo's Data Systems Group, a data system is needed to retain the information of a seminar series and automatically coordinate the tasks while interacting with various users. Such a first-of-its-kind data system can significantly contribute to advancing (research and life at) research groups and institutes.

a Data System for Organizing Vacations (DSOV)

In an organization such as Cheriton School of Computer Science, students and employees go on vacation anytime during their study/work terms with the approval and encouragement of their supervisors and manager. The supervisors and manager have to evaluate different calendars and criteria to decide on the acceptance of vacations. The persons who go on vacation also need to know the best time to go on vacation and the challenges of their absence.

A data system that stores the absence/presence of people and the tasks that run on calendar days, would help provide recommendations and answer queries to decision-makers. E.g., what is the best time to take a vacation, whether a last-minute vacation request would be accepted, what should be done during the absence, who is currently in absence, how often people go on vacation, who has not gone on a vacation in their working year and can get a recommendation for planning a trip, etc.

a Data System for Dental Appointments (DSDA)

A dental clinic such as Campus Dentist has a limited number of dentists, technicians, therapists, hygienists, reception staff, and multiple clients each with different schedules, different dental health conditions, and different insurance plans. Scheduling the appointments for these clients requires access to the working and personal calendars of clients, affordability of expenses, insurance coverage, the availability of a dental team, and ensuring the necessity of an appointment.

A data system that integrates the booked appointments, the availability of clients and the dental team, and the insurance plans, and also recommends the next appointments can help facilitate appointment bookings, reduce expenses for the clinic and clients by avoiding unnecessary appointments and early diagnosis, enhance the productivity of the dental team and clients by spending less time on bookings, and improve dental health and user experience.

a Renovation for Davis Centre (RenoDC)

Many rooms in Davis Centre are in need of renovation with a high potential of transforming the school. For instance, the CS Lounge has broken sofas with dirty and unwelcoming covering materials, microwaves that do not get cleaned and used properly, a small fridge, no dish washing machine, a low variety of seats, etc; The washrooms are not equally and properly equipped and lack minimum standards of a healthy frequently-visited place for kids, all employees and students, etc. The office doors are cracked, and not securely and properly locked requiring maintenance.

Please let me know if you are interested in these projects or would like to recommend a job or opportunity.

. My email address is available on my CV.

. My research details and coordinates are available at Publications page on this site.

Some people wrote recommendations for me on LinkedIn. Let me know, if you would like to have a say too.