Publications

Authorship verification is used to link texts written by the same author without needing a model per author, making it useful for deanonymizing users spreading text with malicious intent. Recent advances in Transformer-based language models hold great promise for author verification, though short context lengths and non-diverse training regimes present challenges for their practical application. In this work, we investigate the effect of these challenges in the application of a Cross-Encoder Transformer-based author verification system under multiple conditions. We perform experiments with four Transformer backbones using differently tuned variants of fanfiction data and found that our BigBird pipeline outperformed Longformer, RoBERTa, and ELECTRA and performed competitively against the official top ranked system from the PAN evaluation. We also examined the effect of authors and fandoms not seen in training on model performance. Through this, we found fandom has the greatest influence on true trials, pairs of text written by the same author, and that a balanced training dataset in terms of class and fandom performed the most consistently.

Global social media networks' omnipresent access, real time responsiveness and ability to connect with and influence people have been responsible for these networks' sweeping growth. However, as an unintended consequence, these defining characteristics helped create a powerful new technology for spread of propaganda and false information. We present a novel approach for characterizing disinformation networks on social media and distinguishing between different network roles using graph embeddings and hierarchical clustering. In addition, using topic filtering, we correlate the node characterization results with proxy opinion estimates.We plan to study opinion dynamics using signal processing on graphs approaches using longer-timescale social media datasets with the goal to model and infer influence among users in social media networks.

The Director's Office at MIT Lincoln Laboratory (MIT LL) requested a comprehensive study on artificial intelligence (AI) focusing on present applications and future science and technology (S&T) opportunities in the Cyber Security and Information Sciences Division (Division 5). This report elaborates on the main results from the study. Since the AI field is evolving so rapidly, the study scope was to look at the recent past and ongoing developments to lead to a set of findings and recommendations. It was important to begin with a short AI history and a lay-of-the-land on representative developments across the Department of Defense (DoD), intelligence communities (IC), and Homeland Security. These areas are addressed in more detail within the report. A main deliverable from the study was to formulate an end-to-end AI canonical architecture that was suitable for a range of applications. The AI canonical architecture, formulated in the study, serves as the guiding framework for all the sections in this report. Even though the study primarily focused on cyber security and information sciences, the enabling technologies are broadly applicable to many other areas. Therefore, we dedicate a full section on enabling technologies in Section 3. The discussion on enabling technologies helps the reader clarify the distinction among AI, machine learning algorithms, and specific techniques to make an end-to-end AI system viable. In order to understand what is the lay-of-the-land in AI, study participants performed a fairly wide reach within MIT LL and external to the Laboratory (government, commercial companies, defense industrial base, peers, academia, and AI centers). In addition to the study participants (shown in the next section under acknowledgements), we also assembled an internal review team (IRT). The IRT was extremely helpful in providing feedback and in helping with the formulation of the study briefings, as we transitioned from datagathering mode to the study synthesis. The format followed throughout the study was to highlight relevant content that substantiates the study findings, and identify a set of recommendations. An important finding is the significant AI investment by the so-called "big 6" commercial companies. These major commercial companies are Google, Amazon, Facebook, Microsoft, Apple, and IBM. They dominate in the AI ecosystem research and development (R&D) investments within the U.S. According to a recent McKinsey Global Institute report, cumulative R&D investment in AI amounts to about $30 billion per year. This amount is substantially higher than the R&D investment within the DoD, IC, and Homeland Security. Therefore, the DoD will need to be very strategic about investing where needed, while at the same time leveraging the technologies already developed and available from a wide range of commercial applications. As we will discuss in Section 1 as part of the AI history, MIT LL has been instrumental in developing advanced AI capabilities. For example, MIT LL has a long history in the development of human language technologies (HLT) by successfully applying machine learning algorithms to difficult problems in speech recognition, machine translation, and speech understanding. Section 4 elaborates on prior applications of these technologies, as well as newer applications in the context of multi-modalities (e.g., speech, text, images, and video). An end-to-end AI system is very well suited to enhancing the capabilities of human language analysis. Section 5 discusses AI's nascent role in cyber security. There have been cases where AI has already provided important benefits. However, much more research is needed in both the application of AI to cyber security and the associated vulnerability to the so-called adversarial AI. Adversarial AI is an area very critical to the DoD, IC, and Homeland Security, where malicious adversaries can disrupt AI systems and make them untrusted in operational environments. This report concludes with specific recommendations by formulating the way forward for Division 5 and a discussion of S&T challenges and opportunities. The S&T challenges and opportunities are centered on the key elements of the AI canonical architecture to strengthen the AI capabilities across the DoD, IC, and Homeland Security in support of national security.

Human trafficking is a form of modern-day slavery affecting an estimated 40 million victims worldwide, primarily through the commercial sexual exploitation of women and children. In the last decade, the advertising of victims has moved from the streets to websites on the Internet, providing greater efficiency and anonymity for sex traffickers. This shift has allowed traffickers to list their victims in multiple geographic areas simultaneously, while also improving operational security by using multiple methods of electronic communication with buyers; complicating the ability of law enforcement to disrupt these illicit organizations. In this paper, we address this issue and present a novel unsupervised and scalable template matching algorithm for analyzing and detecting complex organizations operating on adult service websites. The algorithm uses only the advertisement content to uncover signature patterns in text that are indicative of organized activities and organizational structure. We apply this method to a large corpus of adult service advertisements retrieved from backpage.com, and show that the networks identified through the algorithm match well with surrogate truth data derived from phone number networks in the same corpus. Further exploration of the results show that the proposed method provides deeper insights into the complex structures of sex trafficking organizations, not possible through networks derived from phone numbers alone. This method provides a powerful new capability for law enforcement to more completely identify and gather evidence about trafficking networks and their operations.

We present a new method to bootstrap filter Twitter language ID labels in our dataset for automatic language identification (LID). Our method combines geolocation, original Twitter LID labels, and Amazon Mechanical Turk to resolve missing and unreliable labels. We are the first to compare LID classification performance using the MIRA algorithm and langid.py. We show classifier performance on different versions of our dataset with high accuracy using only Twitter data, without ground truth, and very few training examples. We also show how Platt Scaling can be use to calibrate MIRA classifier output values into a probability distribution over candidate classes, making the output more intuitive. Our method allows for fine-grained distinctions between similar languages and dialects and allows us to rediscover the language composition of our Twitter dataset.

Machine learning methods in Human Language Technology have reached a stage of maturity where widespread use is both possible and desirable. The MIT Lincoln Laboratory LLTools software suite provides a step towards this goal by providing a set of easily accessible frameworks for incorporating speech, text, and entity resolution components into larger applications. For the speech processing component, the pySLGR (Speaker, Language, Gender Recognition) tool provides signal processing, standard feature analysis, speech utterance embedding, and machine learning modeling methods in Python. The text processing component in LLTools extracts semantically meaningful insights from unstructured data via entity extraction, topic modeling, and document classification. The entity resolution component in LLTools provides approximate string matching, author recognition and graph-based methods for identifying and linking different instances of the same real-world entity. We show through two applications that LLTools can be used to rapidly create and train research prototypes for human language processing.

Patent litigation is an expensive and time-consuming process. To minimize its impact on the participants in the patent lifecycle, automatic determination of litigation potential is a compelling machine learning application. In this paper, we consider preliminary methods for the prediction of a patent being involved in litigation using metadata, content, and graph features. Metadata features are top-level easily-extractable features, i.e., assignee, number of claims, etc. The content feature performs lexical analysis of the claims associated to a patent. Graph features use relational learning to summarize patent references. We apply our methods on US patents using a labeled data set. Prior work has focused on metadata-only features, but we show that both graph and content features have significant predictive capability. Additionally, fusing all features results in improved performance. We also perform a preliminary examination of some of the qualitative factors that may have significant importance in patent litigation.

Automatic extraction of intelligent and useful information from data is one of the main goals in data science. Traditional approaches have focused on learning from structured features, i.e., information in a relational database. However, most of the data encountered in practice are unstructured (i.e., social media posts, forums, emails and web logs); they do not have a predefined schema or format. In this work, we examine unsupervised methods for processing unstructured text data, extracting relevant information, and transforming it into structured information that can then be leveraged in various applications such as graph analysis and matching entities across different platforms. Various efforts have been proposed to develop algorithms for processing unstructured text data. At a top level, text can be either summarized by document level features (i.e., language, topic, genre, etc.) or analyzed at a word or sub-word level. Text analytics can be unsupervised, semi-supervised, or supervised. In this work, we focus on word analysis and unsupervised methods. Unsupervised (or semi-supervised) methods require less human annotation and can easily fulfill the role of automatic analysis. For text analysis, we focus on methods for finding relevant words in the text. Specifically, we look at social media data and attempt to predict hashtags for users' posts. The resulting hashtags can be used for downstream processing such as graph analysis. Automatic hashtag annotation is closely related to automatic tag extraction and keyword extraction. Techniques for hashtags extraction include topic analysis, supervised classifiers, machine translation methods, and collaborative filtering. Methods for keyword extraction include graph-based and topical analysis of text.

The automatic determination of emotional state from multimedia content is an inherently challenging problem with a broad range of applications including biomedical diagnostics, multimedia retrieval, and human computer interfaces. The Audio Video Emotion Challenge (AVEC) 2016 provides a well-defined framework for developing and rigorously evaluating innovative approaches for estimating the arousal and valence states of emotion as a function of time. It presents the opportunity for investigating multimodal solutions that include audio, video, and physiological sensor signals. This paper provides an overview of our AVEC Emotion Challenge system, which uses multi-feature learning and fusion across all available modalities. It includes a number of technical contributions, including the development of novel high- and low-level features for modeling emotion in the audio, video, and physiological channels. Low-level features include modeling arousal in audio with minimal prosodic-based descriptors. High-level features are derived from supervised and unsupervised machine learning approaches based on sparse coding and deep learning. Finally, a state space estimation approach is applied for score fusion that demonstrates the importance of exploiting the time-series nature of the arousal and valence states. The resulting system outperforms the baseline systems [10] on the test evaluation set with an achieved Concordant Correlation Coefficient (CCC) for arousal of 0.770 vs 0.702 (baseline) and for valence of 0.687 vs 0.638. Future work will focus on exploiting the time-varying nature of individual channels in the multi-modal framework.

Major depressive disorder (MDD) is known to result in neurophysiological and neurocognitive changes that affect control of motor, linguistic, and cognitive functions. MDD's impact on these processes is reflected in an individual's communication via coupled mechanisms: vocal articulation, facial gesturing and choice of content to convey in a dialogue. In particular, MDD-induced neurophysiological changes are associated with a decline in dynamics and coordination of speech and facial motor control, while neurocognitive changes influence dialogue semantics. In this paper, biomarkers are derived from all of these modalities, drawing first from previously developed neurophysiologically motivated speech and facial coordination and timing features. In addition, a novel indicator of lower vocal tract constriction in articulation is incorporated that relates to vocal projection. Semantic features are analyzed for subject/avatar dialogue content using a sparse coded lexical embedding space, and for contextual clues related to the subject's present or past depression status. The features and depression classification system were developed for the 6th International Audio/Video Emotion Challenge (AVEC), which provides data consisting of audio, video-based facial action units, and transcribed text of individuals communicating with the human-controlled avatar. A clinical Patient Health Questionnaire (PHQ) score and binary depression decision are provided for each participant. PHQ predictions were obtained by fusing outputs from a Gaussian staircase regressor for each feature set, with results on the development set of mean F1=0.81, RMSE=5.31, and MAE=3.34. These compare favorably to the challenge baseline development results of mean F1=0.73, RMSE=6.62, and MAE=5.52. On test set evaluation, our system obtained a mean F1=0.70, which is similar to the challenge baseline test result. Future work calls for consideration of joint feature analyses across modalities in an effort to detect neurological disorders based on the interplay of motor, linguistic, affective, and cognitive components of communication.