Drug similarity has been studied to support downstream clinical tasks such as inferring novel properties of drugs (e.g. side effects, indications, interactions) from known properties. The growing availability of new types of drug features brings the opportunity of learning a more comprehensive and accurate drug similarity that represents the full spectrum of underlying drug relations. However, it is challenging to integrate these heterogeneous, noisy, nonlinear-related information to learn accurate similarity measures especially when labels are scarce. Moreover, there is a trade-off between accuracy and interpretability. In this paper, we propose to learn accurate and interpretable similarity measures from multiple types of drug features. In particular, we model the integration using multi-view graph auto-encoders, and add attentive mechanism to determine the weights for each view with respect to corresponding tasks and features for better interpretability. Our model has flexible design for both semi-supervised and unsupervised settings. Experimental results demonstrated significant predictive accuracy improvement. Case studies also showed better model capacity (e.g. embed node features) and interpretability.

Machine learning and data science require more than just throwing data into a python library and utilizing whatever comes out. Data scientists need to actually understand the data and the processes behind the data to be able to implement a successful system. One key methodology to implementation is knowing when a model might benefit from utilizing bootstrapping methods. These are what are called ensemble models. Some examples of ensemble models are AdaBoost and Stochastic Gradient Boosting. They can help improve algorithm accuracy or improve the robustness of a model.

ROC is a graphic plot illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied. The critical point here is "binary classifier" and "varying threshold". I will show you how to plot ROC for multi-label classifier by the one-vs-all approach as well. AUC is the percentage of this area that is under this ROC curve, ranging between 0 1. ROC is a great way to visualize the performance of a binary classifier, and AUC is one single number to summarize a classifier's performance by assessing the ranking regarding separation of the two classes.

Smart machines are giving storytellers and risk managers alike a helping hand. Burgeoning data analyzed by ever more intelligent machines are opening pathways to surprising applications and providing solutions to problems that have been out of reach. In the film industry, machines "watch" movies and videos, charting their emotional intensity and giving content creators clues about to how to make stories more appealing. And in banking, AI's ability to detect anomalies among millions of transactions helps bank risk officers eliminate false positives that are a drain on productivity. For a growing number of industries, AI is tilting the playing field--you'll need to understand how before your competitors do. Machine-learning models can help screenwriters and directors fine-tune scripts and imagery.

A startup that uses artificial intelligence to help emergency dispatchers identify signs of cardiac arrest over the phone will begin testing its software across Europe this summer. Danish firm Corti says its algorithms can recognize out-of-hospital cardiac arrests (those that occur in the home or public) more quickly and accurately than humans. The software has already been deployed in Copenhagen, but this year, it will start four new pilots in as-yet-unnamed European cities in partnership with the European Emergency Number Association (EENA). Quick recognition of cardiac arrests is vital, as every minute that passes without treatment reduces an individual's chances of survival by 7 to 10 percent. Corti's software works by listening in during emergency calls and looking out for a number of "verbal and non-verbal patterns of communication." These include cues like a caller's tone of voice and whether or not the subject is breathing.

Every year, thousands of people receive consumer product related injuries. Research indicates that online customer reviews can be processed to autonomously identify product safety issues. Early identification of safety issues can lead to earlier recalls, and thus fewer injuries and deaths. A dataset of product reviews from Amazon.com was compiled, along with \emph{SaferProducts.gov} complaints and recall descriptions from the Consumer Product Safety Commission (CPSC) and European Commission Rapid Alert system. A system was built to clean the collected text and to extract relevant features. Dimensionality reduction was performed by computing feature relevance through a Random Forest and discarding features with low information gain. Various classifiers were analyzed, including Logistic Regression, SVMs, Na{\"i}ve-Bayes, Random Forests, and an Ensemble classifier. Experimentation with various features and classifier combinations resulted in a logistic regression model with 70.2\% precision in the top 50 reviews surfaced. This classifier outperforms all benchmarks set by related literature and consumer product safety professionals.

Random Fourier features is one of the most popular techniques for scaling up kernel methods, such as kernel ridge regression. However, despite impressive empirical results, the statistical properties of random Fourier features are still not well understood. In this paper we take steps toward filling this gap. Specifically, we approach random Fourier features from a spectral matrix approximation point of view, give tight bounds on the number of Fourier features required to achieve a spectral approximation, and show how spectral matrix approximation bounds imply statistical guarantees for kernel ridge regression. Qualitatively, our results are twofold: on the one hand, we show that random Fourier feature approximation can provably speed up kernel ridge regression under reasonable assumptions. At the same time, we show that the method is suboptimal, and sampling from a modified distribution in Fourier space, given by the leverage function of the kernel, yields provably better performance. We study this optimal sampling distribution for the Gaussian kernel, achieving a nearly complete characterization for the case of low-dimensional bounded datasets. Based on this characterization, we propose an efficient sampling scheme with guarantees superior to random Fourier features in this regime.

A big problem with supervised machine learning is the need for huge amounts of labeled data. It's a big problem especially if you don't have the labeled data--and even in a world awash with big data, most of us don't. Although a few companies have access to enormous quantities of certain kinds of labeled data, for most organizations and many applications, creating sufficient quantities of the right kind of labeled data is cost prohibitive or impossible. Sometimes the domain is one in which there just isn't much data (for example, when diagnosing a rare disease or determining whether a signature matches a few known exemplars). Other times the volume of data needed multiplied by the cost of human labeling by Amazon Turkers or summer interns is just too high.

Traditional cross-modal retrieval assumes explicit association of concepts across modalities, where there is no ambiguity in how the concepts are linked to each other, e.g., when we do the image search with a query "dogs", we expect to see dog images. In this paper, we consider a different setting for cross-modal retrieval where data from different modalities are implicitly linked via concepts that must be inferred by high-level reasoning; we call this setting implicit concept association. To foster future research in this setting, we present a new dataset containing 47K pairs of animated GIFs and sentences crawled from the web, in which the GIFs depict physical or emotional reactions to the scenarios described in the text (called "reaction GIFs"). We report on a user study showing that, despite the presence of implicit concept association, humans are able to identify video-sentence pairs with matching concepts, suggesting the feasibility of our task. Furthermore, we propose a novel visual-semantic embedding network based on multiple instance learning. Unlike traditional approaches, we compute multiple embeddings from each modality, each representing different concepts, and measure their similarity by considering all possible combinations of visual-semantic embeddings in the framework of multiple instance learning. We evaluate our approach on two video-sentence datasets with explicit and implicit concept association and report competitive results compared to existing approaches on cross-modal retrieval.

Fake news may be intentionally created to promote economic, political and social interests, and can lead to negative impacts on humans beliefs and decisions. Hence, detection of fake news is an emerging problem that has become extremely prevalent during the last few years. Most existing works on this topic focus on manual feature extraction and supervised classification models leveraging a large number of labeled (fake or real) articles. In contrast, we focus on content-based detection of fake news articles, while assuming that we have a small amount of labels, made available by manual fact-checkers or automated sources. We argue this is a more realistic setting in the presence of massive amounts of content, most of which cannot be easily factchecked. To that end, we represent collections of news articles as multi-dimensional tensors, leverage tensor decomposition to derive concise article embeddings that capture spatial/contextual information about each news article, and use those embeddings to create an article-by-article graph on which we propagate limited labels. Results on three real-world datasets show that our method performs on par or better than existing models that are fully supervised, in that we achieve better detection accuracy using fewer labels. In particular, our proposed method achieves 75.43% of accuracy using only 30% of labels of a public dataset while an SVM-based classifier achieved 67.43%. Furthermore, our method achieves 70.92% of accuracy in a large dataset using only 2% of labels.