Joonseok Lee

L-EnsNMF: Boosted Local Topic Discovery via Ensemble of Nonnegative Matrix Factorization

Nonnegative matrix factorization (NMF) has beenwidely applied in many domains. In document analysis, it hasbeen increasingly used in topic modeling applications, where aset of underlying topics are revealed by a low-rank factor matrixfrom NMF. However, it is often the case that the resulting topicsgive only general topic information in the data, which tends notto convey much information. To tackle this problem, we proposea novel ensemble model of nonnegative matrix factorizationfor discovering high-quality local topics. Our method leveragesthe idea of an ensemble model, which has been successfulin supervised learning, into an unsupervised topic modelingcontext. That is, our model successively performs NMF givena residual matrix obtained from previous stages and generatesa sequence of topic sets. Our algorithm for updating the inputmatrix has novelty in two aspects. The first lies in utilizing theresidual matrix inspired by a state-of-the-art gradient boostingmodel, and the second stems from applying a sophisticatedlocal weighting scheme on the given matrix to enhance thelocality of topics, which in turn delivers high-quality, focusedtopics of interest to users. We evaluate our proposed method bycomparing it against other topic modeling methods, such as afew variants of NMF and latent Dirichlet allocation, in termsof various evaluation measures representing topic coherence, diversity, coverage, computing time, and so on. We also presentqualitative evaluation on the topics discovered by our methodusing several real-world data sets.

Leveraging Knowledge Bases for Contextual Entity Exploration

Users today are constantly switching back and forth from applications where they consume or create content (such as e-books and productivity suites like Microsoft Office and Google Docs) to search engines where they satisfy their information needs. Unfortunately, though, this leads to a suboptimal user experience as the search engine lacks any knowledge about the content that the user is authoring or consuming in the application. As a result, productivity suites are starting to incorporate features that let the user explore while they work. Existing work in the literature that can be applied to this problem takes a standard bag-of-words information retrieval approach, which consists of automatically creating a query that includes not only the target phrase or entity chosen by the user but also relevant terms from the context. While these approaches have been successful, they are inherently limited to returning results (documents) that have a syntactic match with the keywords in the query. We argue that the limitations of these approaches can be overcome by leveraging semantic signals from a knowledge graph built from knowledge bases such as Wikipedia. We present a system called Lewis for retrieving contextually relevant entity results leveraging a knowledge graph, and perform a large scale crowdsourcing experiment in the context of an e-reader scenario, which shows that Lewis can outperform the state-of-the-art contextual entity recommendation systems by more than 20% in terms of the MAP score.

Local Topic Discovery via Boosted Ensemble of Nonnegative Matrix Factorization

Nonnegative matrix factorization (NMF) has been increasingly popular for topic modeling of largescale documents. However, the resulting topics often represent only general, thus redundant information about the data rather than minor, but potentially meaningful information to users. To tackle this problem, we propose a novel ensemble model of nonnegative matrix factorization for discovering high-quality local topics. Our method leverages the idea of an ensemble model to successively perform NMF given a residual matrix obtained from previous stages and generates a sequence of topic sets. The novelty of our method lies in the fact that it utilizes the residual matrix inspired by a state-of-theart gradient boosting model and applies a sophisticated local weighting scheme on the given matrix to enhance the locality of topics, which in turn delivers high-quality, focused topics of interest to users.1

Collaborative Deep Metric Learning for Video Understanding

The goal of video understanding is to develop algorithms that enable machines understand videos at the level of human experts. Researchers have tackled various domains including video classification, search, personalized recommendation, and more. However, there is a research gap in combining these domains in one unified learning framework. Towards that, we propose a deep network that embeds videos using their audio-visual content, onto a metric space which preserves video-to-video relationships. Then, we use the trained embedding network to tackle various domains including video classification and recommendation, showing significant improvements over state-of-the-art baselines. The proposed approach is highly scalable to deploy on large-scale video sharing platforms like YouTube.

Localized user-driven topic discovery via boosted ensemble of nonnegative matrix factorization

Nonnegative matrix factorization (NMF) has been widely used in topic modeling of large-scale document corpora, where a set of underlying topics are extracted by a low-rank factor matrix from NMF. However, the resulting topics often convey only general, thus redundant information about the documents rather than information that might be minor, but potentially meaningful to users. To address this problem, we present a novel ensemble method based on nonnegative matrix factorization that discovers meaningful local topics. Our method leverages the idea of an ensemble model, which has shown advantages in supervised learning, into an unsupervised topic modeling context. That is, our model successively performs NMF given a residual matrix obtained from previous stages and generates a sequence of topic sets. The algorithm we employ to update is novel in two aspects. The first lies in utilizing the residual matrix inspired by a state-of-the-art gradient boosting model, and the second stems from applying a sophisticated local weighting scheme on the given matrix to enhance the locality of topics, which in turn delivers high-quality, focused topics of interest to users. We subsequently extend this ensemble model by adding keyword- and document-based user interaction to introduce user-driven topic discovery.