We present the evolution of a large-scale content recommendation platform for LinkedIn Learning, serving 645M+ LinkedIn users across several different channels (e.g., desktop, mobile). We address challenges and complexities from both algorithms and infrastructure perspectives. We describe the progression from unsupervised models that exploit member similarity with course content, to supervised learning models leveraging member interactions with courses, and finally to hyper-personalized mixed-effects models with several million coefficients. For all the experiments, we include metric lifts achieved via online A/B tests and illustrate the trade-offs between computation and storage requirements.

Keyboard layout errors and homoglyphs in cross-language queries impact our abil-ity to correctly interpret user informa-tion needs and offer relevant results. We present a machine learning approach to correcting these errors, based largely on character-level n-gram features. We demonstrate superior performance over rule-based methods, as well as a signif-icant reduction in the number of queries that yield null search results. 1

We investigate how the affordances provided by alternative interfaces for on-line discussion forums affect the structure of the discourse that unfolds. In order to investigate this impact, we compare the predictive power of time related and text similarity related features for identifying parent-child links between messages. The results from this work using this methodology suggest that interfaces that make parent-child relationships between messages explicit and do not constrain the choice of previous messages that users can reply to allow patterns of conversational behavior that violate the assumptions of traditional, tree-structured models of discourse where time related and similarity related features are highly predictive. An implication for future work is that because there is evidence that interface affordances affect the form of conversational contributions, techniques that process on-line communication data may need to be adapted for different communication interfaces.

Health activity monitoring (HAM) has received increasing attention due to the rapid advances of both hardware and software technologies and strong environmental and public health needs. In this paper, we describe the architecture and implementation of the Epi-SPIRE prototype, which is a novel health activity monitoring system that generates alerts from environmental, behavioral, and public health data sources. A model-based approach is used to develop disease and behavior models from multi-modal heterogeneous data sources. Furthermore, a model-based indexing technique has been developed to speed up the data access and retrieval. This system has been successfully applied to various genuine and simulated diseases outbreaks scenarios'.