Derek Doran

Ning Xie, Farley Lai, Derek Doran et al.

We introduce a new inference task - Visual Entailment (VE) - which differs from traditional Textual Entailment (TE) tasks whereby a premise is defined by an image, rather than a natural language sentence as in TE tasks. A novel dataset SNLI-VE (publicly available at https://github.com/necla-ml/SNLI-VE) is proposed for VE tasks based on the Stanford Natural Language Inference corpus and Flickr30k. We introduce a differentiable architecture called the Explainable Visual Entailment model (EVE) to tackle the VE problem. EVE and several other state-of-the-art visual question answering (VQA) based models are evaluated on the SNLI-VE dataset, facilitating grounded language understanding and providing insights on how modern VQA based models perform.

Gabrielle Ras, Ning Xie, Marcel van Gerven et al.

Deep neural networks (DNNs) have become a proven and indispensable machine learning tool. As a black-box model, it remains difficult to diagnose what aspects of the model's input drive the decisions of a DNN. In countless real-world domains, from legislation and law enforcement to healthcare, such diagnosis is essential to ensure that DNN decisions are driven by aspects appropriate in the context of its use. The development of methods and studies enabling the explanation of a DNN's decisions has thus blossomed into an active, broad area of research. A practitioner wanting to study explainable deep learning may be intimidated by the plethora of orthogonal directions the field has taken. This complexity is further exacerbated by competing definitions of what it means ``to explain'' the actions of a DNN and to evaluate an approach's ``ability to explain''. This article offers a field guide to explore the space of explainable deep learning aimed at those uninitiated in the field. The field guide: i) Introduces three simple dimensions defining the space of foundational methods that contribute to explainable deep learning, ii) discusses the evaluations for model explanations, iii) places explainability in the context of other related deep learning research areas, and iv) finally elaborates on user-oriented explanation designing and potential future directions on explainable deep learning. We hope the guide is used as an easy-to-digest starting point for those just embarking on research in this field.

Farley Lai, Ning Xie, Derek Doran et al.

In this paper, we introduce a contextual grounding approach that captures the context in corresponding text entities and image regions to improve the grounding accuracy. Specifically, the proposed architecture accepts pre-trained text token embeddings and image object features from an off-the-shelf object detector as input. Additional encoding to capture the positional and spatial information can be added to enhance the feature quality. There are separate text and image branches facilitating respective architectural refinements for different modalities. The text branch is pre-trained on a large-scale masked language modeling task while the image branch is trained from scratch. Next, the model learns the contextual representations of the text tokens and image objects through layers of high-order interaction respectively. The final grounding head ranks the correspondence between the textual and visual representations through cross-modal interaction. In the evaluation, we show that our model achieves the state-of-the-art grounding accuracy of 71.36% over the Flickr30K Entities dataset. No additional pre-training is necessary to deliver competitive results compared with related work that often requires task-agnostic and task-specific pre-training on cross-modal dadasets. The implementation is publicly available at https://gitlab.com/necla-ml/grounding.

Mahdieh Zabihimayvan, Derek Doran

Phishing as one of the most well-known cybercrime activities is a deception of online users to steal their personal or confidential information by impersonating a legitimate website. Several machine learning-based strategies have been proposed to detect phishing websites. These techniques are dependent on the features extracted from the website samples. However, few studies have actually considered efficient feature selection for detecting phishing attacks. In this work, we investigate an agreement on the definitive features which should be used in phishing detection. We apply Fuzzy Rough Set (FRS) theory as a tool to select most effective features from three benchmarked data sets. The selected features are fed into three often used classifiers for phishing detection. To evaluate the FRS feature selection in developing a generalizable phishing detection, the classifiers are trained by a separate out-of-sample data set of 14,000 website samples. The maximum F-measure gained by FRS feature selection is 95% using Random Forest classification. Also, there are 9 universal features selected by FRS over all the three data sets. The F-measure value using this universal feature set is approximately 93% which is a comparable result in contrast to the FRS performance. Since the universal feature set contains no features from third-part services, this finding implies that with no inquiry from external sources, we can gain a faster phishing detection which is also robust toward zero-day attacks.

Ning Xie, Farley Lai, Derek Doran et al.

Existing visual reasoning datasets such as Visual Question Answering (VQA), often suffer from biases conditioned on the question, image or answer distributions. The recently proposed CLEVR dataset addresses these limitations and requires fine-grained reasoning but the dataset is synthetic and consists of similar objects and sentence structures across the dataset. In this paper, we introduce a new inference task, Visual Entailment (VE) - consisting of image-sentence pairs whereby a premise is defined by an image, rather than a natural language sentence as in traditional Textual Entailment tasks. The goal of a trained VE model is to predict whether the image semantically entails the text. To realize this task, we build a dataset SNLI-VE based on the Stanford Natural Language Inference corpus and Flickr30k dataset. We evaluate various existing VQA baselines and build a model called Explainable Visual Entailment (EVE) system to address the VE task. EVE achieves up to 71% accuracy and outperforms several other state-of-the-art VQA based models. Finally, we demonstrate the explainability of EVE through cross-modal attention visualizations. The SNLI-VE dataset is publicly available at https://github.com/ necla-ml/SNLI-VE.

Kyle Brown, Derek Doran, Ryan Kramer et al.

Strong regulations in the financial industry mean that any decisions based on machine learning need to be explained. This precludes the use of powerful supervised techniques such as neural networks. In this study we propose a new unsupervised and semi-supervised technique known as the topological hierarchical decomposition (THD). This process breaks a dataset down into ever smaller groups, where groups are associated with a simplicial complex that approximate the underlying topology of a dataset. We apply THD to the FICO machine learning challenge dataset, consisting of anonymized home equity loan applications using the MAPPER algorithm to build simplicial complexes. We identify different groups of individuals unable to pay back loans, and illustrate how the distribution of feature values in a simplicial complex can be used to explain the decision to grant or deny a loan by extracting illustrative explanations from two THDs on the dataset.

Monireh Ebrahimi, Md Kamruzzaman Sarker, Federico Bianchi et al.

Semantic Web knowledge representation standards, and in particular RDF and OWL, often come endowed with a formal semantics which is considered to be of fundamental importance for the field. Reasoning, i.e., the drawing of logical inferences from knowledge expressed in such standards, is traditionally based on logical deductive methods and algorithms which can be proven to be sound and complete and terminating, i.e. correct in a very strong sense. For various reasons, though, in particular, the scalability issues arising from the ever-increasing amounts of Semantic Web data available and the inability of deductive algorithms to deal with noise in the data, it has been argued that alternative means of reasoning should be investigated which bear high promise for high scalability and better robustness. From this perspective, deductive algorithms can be considered the gold standard regarding correctness against which alternative methods need to be tested. In this paper, we show that it is possible to train a Deep Learning system on RDF knowledge graphs, such that it is able to perform reasoning over new RDF knowledge graphs, with high precision and recall compared to the deductive gold standard.

Giuseppe Nebbione, Derek Doran, Srikanth Nadella et al.

There are many examples of 'wisdom of the crowd' effects in which the large number of participants imparts confidence in the collective judgment of the crowd. But how do we form an aggregated judgment when the size of the crowd is limited? Whose judgments do we include, and whose do we accord the most weight? This paper considers this problem in the context of geopolitical event forecasting, where volunteer analysts are queried to give their expertise, confidence, and predictions about the outcome of an event. We develop a forecast aggregation model that integrates topical information about a question, meta-data about a pair of forecasters, and their predictions in a deep siamese neural network that decides which forecasters' predictions are more likely to be close to the correct response. A ranking of the forecasters is induced from a tournament of pair-wise forecaster comparisons, with the ranking used to create an aggregate forecast. Preliminary results find the aggregate prediction of the best forecasters ranked by our deep siamese network model consistently beats typical aggregation techniques by Brier score.

Kyle Brown, Derek Doran

Critical to evaluating the capacity, scalability, and availability of web systems are realistic web traffic generators. Web traffic generation is a classic research problem, no generator accounts for the characteristics of web robots or crawlers that are now the dominant source of traffic to a web server. Administrators are thus unable to test, stress, and evaluate how their systems perform in the face of ever increasing levels of web robot traffic. To resolve this problem, this paper introduces a novel approach to generate synthetic web robot traffic with high fidelity. It generates traffic that accounts for both the temporal and behavioral qualities of robot traffic by statistical and Bayesian models that are fitted to the properties of robot traffic seen in web logs from North America and Europe. We evaluate our traffic generator by comparing the characteristics of generated traffic to those of the original data. We look at session arrival rates, inter-arrival times and session lengths, comparing and contrasting them between generated and real traffic. Finally, we show that our generated traffic affects cache performance similarly to actual traffic, using the common LRU and LFU eviction policies.

Matthew Piekenbrock, Derek Doran

This paper presents a framework for intrinsic point of interest discovery from trajectory databases. Intrinsic points of interest are regions of a geospatial area innately defined by the spatial and temporal aspects of trajectory data, and can be of varying size, shape, and resolution. Any trajectory database exhibits such points of interest, and hence are intrinsic, as compared to most other point of interest definitions which are said to be extrinsic, as they require trajectory metadata, external knowledge about the region the trajectories are observed, or other application-specific information. Spatial and temporal aspects are qualities of any trajectory database, making the framework applicable to data from any domain and of any resolution. The framework is developed under recent developments on the consistency of nonparametric hierarchical density estimators and enables the possibility of formal statistical inference and evaluation over such intrinsic points of interest. Comparisons of the POIs uncovered by the framework in synthetic truth data to thousands of parameter settings for common POI discovery methods show a marked improvement in fidelity without the need to tune any parameters by hand.

Ning Xie, Md Kamruzzaman Sarker, Derek Doran et al.

Many current methods to interpret convolutional neural networks (CNNs) use visualization techniques and words to highlight concepts of the input seemingly relevant to a CNN's decision. The methods hypothesize that the recognition of these concepts are instrumental in the decision a CNN reaches, but the nature of this relationship has not been well explored. To address this gap, this paper examines the quality of a concept's recognition by a CNN and the degree to which the recognitions are associated with CNN decisions. The study considers a CNN trained for scene recognition over the ADE20k dataset. It uses a novel approach to find and score the strength of minimally distributed representations of input concepts (defined by objects in scene images) across late stage feature maps. Subsequent analysis finds evidence that concept recognition impacts decision making. Strong recognition of concepts frequently-occurring in few scenes are indicative of correct decisions, but recognizing concepts common to many scenes may mislead the network.

Md Kamruzzaman Sarker, Ning Xie, Derek Doran et al.

The ever increasing prevalence of publicly available structured data on the World Wide Web enables new applications in a variety of domains. In this paper, we provide a conceptual approach that leverages such data in order to explain the input-output behavior of trained artificial neural networks. We apply existing Semantic Web technologies in order to provide an experimental proof of concept.

Derek Doran, Sarah Schulz, Tarek R. Besold

We characterize three notions of explainable AI that cut across research fields: opaque systems that offer no insight into its algo- rithmic mechanisms; interpretable systems where users can mathemat- ically analyze its algorithmic mechanisms; and comprehensible systems that emit symbols enabling user-driven explanations of how a conclusion is reached. The paper is motivated by a corpus analysis of NIPS, ACL, COGSCI, and ICCV/ECCV paper titles showing differences in how work on explainable AI is positioned in various fields. We close by introducing a fourth notion: truly explainable systems, where automated reasoning is central to output crafted explanations without requiring human post processing as final step of the generative process.

Sanjaya Wijeratne, Lakshika Balasuriya, Amit Sheth et al.

Emoji have grown to become one of the most important forms of communication on the web. With its widespread use, measuring the similarity of emoji has become an important problem for contemporary text processing since it lies at the heart of sentiment analysis, search, and interface design tasks. This paper presents a comprehensive analysis of the semantic similarity of emoji through embedding models that are learned over machine-readable emoji meanings in the EmojiNet knowledge base. Using emoji descriptions, emoji sense labels and emoji sense definitions, and with different training corpora obtained from Twitter and Google News, we develop and test multiple embedding models to measure emoji similarity. To evaluate our work, we create a new dataset called EmoSim508, which assigns human-annotated semantic similarity scores to a set of 508 carefully selected emoji pairs. After validation with EmoSim508, we present a real-world use-case of our emoji embedding models using a sentiment analysis task and show that our models outperform the previous best-performing emoji embedding model on this task. The EmoSim508 dataset and our emoji embedding models are publicly released with this paper and can be downloaded from http://emojinet.knoesis.org/.

Sanjaya Wijeratne, Lakshika Balasuriya, Amit Sheth et al.

This paper presents the release of EmojiNet, the largest machine-readable emoji sense inventory that links Unicode emoji representations to their English meanings extracted from the Web. EmojiNet is a dataset consisting of: (i) 12,904 sense labels over 2,389 emoji, which were extracted from the web and linked to machine-readable sense definitions seen in BabelNet, (ii) context words associated with each emoji sense, which are inferred through word embedding models trained over Google News corpus and a Twitter message corpus for each emoji sense definition, and (iii) recognizing discrepancies in the presentation of emoji on different platforms, specification of the most likely platform-based emoji sense for a selected set of emoji. The dataset is hosted as an open service with a REST API and is available at http://emojinet.knoesis.org/. The development of this dataset, evaluation of its quality, and its applications including emoji sense disambiguation and emoji sense similarity are discussed.

Lakshika Balasuriya, Sanjaya Wijeratne, Derek Doran et al.

Most street gang members use Twitter to intimidate others, to present outrageous images and statements to the world, and to share recent illegal activities. Their tweets may thus be useful to law enforcement agencies to discover clues about recent crimes or to anticipate ones that may occur. Finding these posts, however, requires a method to discover gang member Twitter profiles. This is a challenging task since gang members represent a very small population of the 320 million Twitter users. This paper studies the problem of automatically finding gang members on Twitter. It outlines a process to curate one of the largest sets of verifiable gang member profiles that have ever been studied. A review of these profiles establishes differences in the language, images, YouTube links, and emojis gang members use compared to the rest of the Twitter population. Features from this review are used to train a series of supervised classifiers. Our classifier achieves a promising F1 score with a low false positive rate.

Sanjaya Wijeratne, Lakshika Balasuriya, Derek Doran et al.

Gang affiliates have joined the masses who use social media to share thoughts and actions publicly. Interestingly, they use this public medium to express recent illegal actions, to intimidate others, and to share outrageous images and statements. Agencies able to unearth these profiles may thus be able to anticipate, stop, or hasten the investigation of gang-related crimes. This paper investigates the use of word embeddings to help identify gang members on Twitter. Building on our previous work, we generate word embeddings that translate what Twitter users post in their profile descriptions, tweets, profile images, and linked YouTube content to a real vector format amenable for machine learning classification. Our experimental results show that pre-trained word embeddings can boost the accuracy of supervised learning algorithms trained over gang members social media posts.

Sanjaya Wijeratne, Lakshika Balasuriya, Amit Sheth et al.

Emoji are a contemporary and extremely popular way to enhance electronic communication. Without rigid semantics attached to them, emoji symbols take on different meanings based on the context of a message. Thus, like the word sense disambiguation task in natural language processing, machines also need to disambiguate the meaning or sense of an emoji. In a first step toward achieving this goal, this paper presents EmojiNet, the first machine readable sense inventory for emoji. EmojiNet is a resource enabling systems to link emoji with their context-specific meaning. It is automatically constructed by integrating multiple emoji resources with BabelNet, which is the most comprehensive multilingual sense inventory available to date. The paper discusses its construction, evaluates the automatic resource creation process, and presents a use case where EmojiNet disambiguates emoji usage in tweets. EmojiNet is available online for use at http://emojinet.knoesis.org.

Derek Doran, Swapna Gokhale, Aldo Dagnino

Associating geo-coordinates with the content of social media posts can enhance many existing applications and services and enable a host of new ones. Unfortunately, a majority of social media posts are not tagged with geo-coordinates. Even when location data is available, it may be inaccurate, very broad or sometimes fictitious. Contemporary location estimation approaches based on analyzing the content of these posts can identify only broad areas such as a city, which limits their usefulness. To address these shortcomings, this paper proposes a methodology to narrowly estimate the geo-coordinates of social media posts with high accuracy. The methodology relies solely on the content of these posts and prior knowledge of the wide geographical region from where the posts originate. An ensemble of language models, which are smoothed over non-overlapping sub-regions of a wider region, lie at the heart of the methodology. Experimental evaluation using a corpus of over half a million tweets from New York City shows that the approach, on an average, estimates locations of tweets to within just 2.15km of their actual positions.