Minsuk Kahng

Random walk based entity ranking on graph for multidimensional recommendation

In many applications, flexibility of recommendation, which is the capability of handling multiple dimensions and various recommendation types, is very important. In this paper, we focus on the flexibility of recommendation and propose a graph-based multidimensional recommendation method. We consider the problem as an entity ranking problem on the graph which is constructed using an implicit feedback dataset (e.g. music listening log), and we adapt Personalized PageRank algorithm to rank entities according to a given query that is represented as a set of entities in the graph. Our model has advantages in that not only can it support the flexibility, but also it can take advantage of exploiting indirect relationships in the graph so that it can perform competitively with the other existing recommendation methods without suffering from the sparsity problem.

Exploiting Contextual Information from Event Logs for Personalized Recommendation

Nowadays, recommender systems are widely used in various domains to help customers access to more satisfying products or services. It is expected that exploiting customers’ contextual information can improve the quality of recommendation results. Most earlier researchers assume that they already have customers’ explicit ratings on items and each rating has customer’s abstracted context (e.g. summer, morning). However, in practical applications, it is not easy to obtain customers’ explicit ratings and their abstract-level contexts. We aim to acquire customers’ preferences and their context by exploiting the information implied in the customers’ previous event logs and to adopt them into a well known recommendation technique, Collaborative Filtering (CF). In this paper, we show how to obtain customers’ implicit preferences from event logs and present a strategy to abstract context information from event logs considering fuzziness in context. In addition, we present several methods to cooperate achieved contextual information and preferences into CF. To evaluate and compare our methods, we conducted several empirical experiments using a set of music listening logs obtained from last.fm, and the results indicate that our methods can improve the quality of recommendation.

Scalable graph exploration and visualization: Sensemaking challenges and opportunities

Making sense of large graph datasets is a fundamental and challenging process that advances science, education and technology. We survey research on graph exploration and visualization approaches aimed at addressing this challenge. Different from existing surveys, our investigation highlights approaches that have strong potential in handling large graphs, algorithmically, visually, or interactively; we also explicitly connect relevant works from multiple research fields - data mining, machine learning, human-computer ineraction, information visualization, information retrieval, and recommender systems - to underline their parallel and complementary contributions to graph sensemaking. We ground our discussion in sensemaking research; we propose a new graph sensemaking hierarchy that categorizes tools and techniques based on how they operate on the graph data (e.g., local vs global). We summarize and compare their strengths and weaknesses, and highlight open challenges. We conclude with future research directions for graph sensemaking.

PathRank: Ranking nodes on a heterogeneous graph for flexible hybrid recommender systems

We present a flexible hybrid recommender system that can emulate collaborative-filtering, Content-based Filtering, context-aware recommendation, and combinations of any of these recommendation semantics. The recommendation problem is modeled as a problem of finding the most relevant nodes for a given set of query nodes on a heterogeneous graph. However, existing node ranking measures cannot fully exploit the semantics behind the different types of nodes and edges in a heterogeneous graph. To overcome the limitation, we present a novel random walk based node ranking measure, PathRank, by extending the Personalized PageRank algorithm. The proposed measure can produce node ranking results with varying semantics by discriminating the different paths on a heterogeneous graph. The experimental results show that our method can produce more diverse and effective recommendation results compared to existing approaches.

A cti V is : Visual Exploration of Industry-Scale Deep Neural Network Models

While deep learning models have achieved state-of-the-art accuracies for many prediction tasks, understanding these models remains a challenge. Despite the recent interest in developing visual tools to help users interpret deep learning models, the complexity and wide variety of models deployed in industry, and the large-scale datasets that they used, pose unique design challenges that are inadequately addressed by existing work. Through participatory design sessions with over 15 researchers and engineers at Facebook, we have developed, deployed, and iteratively improved ActiVis, an interactive visualization system for interpreting large-scale deep learning models and results. By tightly integrating multiple coordinated views, such as a computation graph overview of the model architecture, and a neuron activation view for pattern discovery and comparison, users can explore complex deep neural network models at both the instance-and subset-level. ActiVis has been deployed on Facebooks machine learning platform. We present case studies with Facebook researchers and engineers, and usage scenarios of how ActiVis may work with different models.

Understanding variations in pediatric asthma care processes in the emergency department using visual analytics

Health care delivery processes consist of complex activity sequences spanning organizational, spatial, and temporal boundaries. Care is human-directed so these processes can have wide variations in cost, quality, and outcome making systemic care process analysis, conformance testing, and improvement challenging. We designed and developed an interactive visual analytic process exploration and discovery tool and used it to explore clinical data from 5784 pediatric asthma emergency department patients.

PathRank: a novel node ranking measure on a heterogeneous graph for recommender systems

In this paper, we present a novel random-walk based node ranking measure, PathRank, which is defined on a heterogeneous graph by extending the Personalized PageRank algorithm. Not only can our proposed measure exploit the semantics behind the different types of nodes and edges in a heterogeneous graph, but also it can emulate various recommendation semantics such as collaborative filtering, content-based filtering, and their combinations. The experimental results show that PathRank can produce more various and effective recommendation results compared to existing approaches.

Ranking objects by following paths in entity-relationship graphs

In this paper, we propose an object ranking method for search and recommendation. By selecting schema-level paths and following them in an entity-relationship graph, it can incorporate diverse semantics existing in the graph. Utilizing this kind of graph-based data models has been recognized as a reasonable way for dealing with heterogeneous data. However, previous work on ranking models using graphs has some limitations. In order to utilize a variety of semantics in multiple types of data, we define a schema path as a basic component of the ranking model. By following the path or a combination of paths, relevant objects could be retrieved or recommended. We present some preliminary experiments to evaluate our method. In addition, we discuss several interesting challenges that can be considered in future work.

GLO-STIX: Graph-Level Operations for Specifying Techniques and Interactive eXploration

The field of graph visualization has produced a wealth of visualization techniques for accomplishing a variety of analysis tasks. Therefore analysts often rely on a suite of different techniques, and visual graph analysis application builders strive to provide this breadth of techniques. To provide a holistic model for specifying network visualization techniques (as opposed to considering each technique in isolation) we present the Graph-Level Operations (GLO) model. We describe a method for identifying GLOs and apply it to identify five classes of GLOs, which can be flexibly combined to re-create six canonical graph visualization techniques. We discuss advantages of the GLO model, including potentially discovering new, effective network visualization techniques and easing the engineering challenges of building multi-technique graph visualization applications. Finally, we implement the GLOs that we identified into the GLO-STIX prototype system that enables an analyst to interactively explore a graph by applying GLOs.

FACETS: Adaptive Local Exploration of Large Graphs

Visualization is a powerful paradigm for exploratory data analysis. Visualizing large graphs, however, often results in excessive edges crossings and overlapping nodes. We propose a new scalable approach called FACETS that helps users adaptively explore large million-node graphs from a local perspective, guiding them to focus on nodes and neighborhoods that are most subjectively interesting to users. We contribute novel ideas to measure this interestingness in terms of how surprising a neighborhood is given the background distribution, as well as how well it matches what the user has chosen to explore. FACETS uses Jensen-Shannon divergence over information-theoretically optimized histograms to calculate the subjective user interest and surprise scores. Participants in a user study found FACETS easy to use, easy to learn, and exciting to use. Empirical runtime analyses demonstrated FACETS’s practical scalability on large real-world graphs with up to 5 million edges, returning results in fewer than 1.5 seconds.