Digital dating abuse is a form of intimate partner violence that uses technology as a medium to propagate fear and cause harm for dating partners. Over several years digital dating abuse has been on the rise, and particularly during COVID-19, the issue has risen exponentially. This project aims to create a tool that raises awareness and detects digital dating from text messages. Previously, we generated a dataset with expert labelers to use supervised machine learning algorithms for abuse detection. However, the cost and time associated with generating human-annotated datasets limit the size of these verified datasets. This poster explores using machine learning algorithms trained on human-annotated datasets to label more extensive crowd-sourced datasets and generate a larger training dataset for abuse detection algorithms. We used Naive Bayes, Decision Tree, LSVM, and LSTM to test for accuracy and speed of labeling this more extensive dataset.

In this paper, we consider the problem of reconstructing trees from traces in the tree edit distance model. Previous work by Davies et al. [10] analyzed special cases of reconstructing labeled trees. In this work, we significantly expand our understanding of this problem by giving general results in the case of arbitrary trees. Namely, we give: a reduction from the tree trace reconstruction problem to the more classical string reconstruction problem when the tree topology is known, a lower bound for learning arbitrary tree topologies, and a general algorithm for learning the topology of any tree using techniques of Nazarov and Peres [19]. We conclude by discussing why arbitrary trees require exponentially many samples under the left propagation model.

In this paper we investigate the problem of learning evolving concepts over a combinatorial structure. Previous work by Emamjomeh-Zadeh et al. [2020] introduced dynamics into interactive learning as a way to model non-static user preferences in clustering problems or recommender systems. We provide many useful contributions to this problem. First, we give a framework that captures both of the models analyzed by [Emamjomeh-Zadeh et al., 2020], which allows us to study any type of concept evolution and matches the same query complexity bounds and running time guarantees of the previous models. Using this general model we solve the open problem of closing the gap between the upper and lower bounds on query complexity. Finally, we study an efficient algorithm where the learner simply follows the feedback at each round, and we provide mistake bounds for low diameter graphs such as cliques, stars, and general o(log n) diameter graphs by using a Markov Chain model.