Vlad Borisov

Slac uses the system of symbolic logic known as Sentential or Propositional Logic, and evaulates the truth-value for any given proposition.

Slac supports unary & binary operators (conjunction, disjunction, negation, conditional, biconditional) and allows for sentences of infinite length and complexity.

Starting from a basic user-input string expression and the truth-value assignment of its atomic components, Slac:

• cleans the string up for internal use
• converts the proposition to reverse-Polish (postfix) notation
• tokenizes the proposition and converts it into a vector of custom class objects
• allows the user to define the assigned truth-values of individual sentences
• evaluates the proposition

CBSS is a job management simulator that applies four distinct scheduling algorithms (FC-FS, SJ-F, SRT-F, & Round Robin) to generated sets of processes, and uses various metrics to compare performance.

CBSS accepts a maximum job boundary as well as a quantum (for use in Round Robin), and randomly generates a fleshed set of processes; all four algorithms are simulated with that set.

Average waiting time is used as the primary metric of algorithm performance comparison, as other metrics (response time, turnaround time, throughput, etc) rely heavily on the presence of I/O or other external factors which would be outside of the scope of the simulation. Developed originally for an OS course with my longtime friend and frequent collaborator, Benjamin Karasik.

MLHD is built off of a CNN model utilizing convolutional & pooling layers and trained on E-MNIST data. The model itself is set up mostly in Python; Tensorflow 2 handles the ML backend w/ tight Keras integration, and JS & React handle user interaction.