November 14th, 2017
ACM Tech Talk Series
A Practitioner’s Introduction to Deep Learning
Ashwin Kumar Ganesan, PhD student
1:00-2:00pm Friday, 17 November 2017?, ITE325, UMBC
In recent years, Deep Neural Networks have been highly successful at performing a number of tasks in computer vision, natural language processing and artificial intelligence in general. The remarkable performance gains have led to universities and industries investing heavily in this space. This investment creates a thriving open source ecosystem of tools & libraries that aid the design of new architectures, algorithm research as well as data collection.
This talk (and hands-on session) introduce people to some of the basics of machine learning, neural networks and discusses some of the popular neural network architectures. We take a dive into one of the popular libraries, Tensorflow, and an associated abstraction library Keras.
To participate in the hands-on aspects of the workshop, bring a laptop computer with Python installed and install the following libraries using pip. For windows or (any other OS) consider doing an installation of anaconda that has all the necessary libraries.
- numpy, scipy & scikit-learn
- tensorflow / tensoflow-gpu (The first one is the GPU version)
- matplotlib for visualizations (if necessary)
- jupyter & ipython (We will use python2.7 in our experiments)
Following are helpful links:
Contact Nisha Pillai (NPillai1 at umbc.edu) with any questions regarding this event.
October 22nd, 2017
Multi-observable Session Reputation Scoring System
11:00-12:00 Monday, 23 October 2017, ITE 346
With increasing adoption of Cloud Computing, cyber attacks have become one of the most effective means for adversaries to inflict damage. To overcome limitations of existing blacklists and whitelists, our research focuses to develop a dynamic reputation scoring model for sessions based on a variety of observable and derived attributes of network traffic. Here we propose a technique to greylist sessions using observables like IP, Domain, URL and File Hash by scoring them numerically based on the events in the session. This enables automatic labeling of possible malicious hosts or users that can help in enriching the existing whitelists or blacklists.
September 10th, 2017
Context-Dependent Privacy and Security Management on Mobile Devices
There are ongoing security and privacy concerns regarding mobile platforms that are being used by a growing number of citizens. Security and privacy models typically used by mobile platforms use one-time permission acquisition mechanisms. However, modifying access rights after initial authorization in mobile systems is often too tedious and complicated for users. User studies show that a typical user does not understand permissions requested by applications or are too eager to use the applications to care to understand the permission implications. For example, the Brightest Flashlight application was reported to have logged precise locations and unique user identifiers, which have nothing to do with a flashlight application’s intended functionality, but more than 50 million users used a version of this application which would have forced them to allow this permission. Given the penetration of mobile devices into our lives, a fine-grained context-dependent security and privacy control approach needs to be created.
We have created Mithril as an end-to-end mobile access control framework that allows us to capture access control needs for specific users, by observing violations of known policies. The framework studies mobile application executables to better inform users of the risks associated with using certain applications. The policy capture process involves an iterative user feedback process that captures policy modifications required to mediate observed violations. Precision of policy is used to determine convergence of the policy capture process. Policy rules in the system are written using Semantic Web technologies and the Platys ontology to define a hierarchical notion of context. Policy rule antecedents are comprised of context elements derived using the Platys ontology employing a query engine, an inference mechanism and mobile sensors. We performed a user study that proves the feasibility of using our violation driven policy capture process to gather user-specific policy modifications.
We contribute to the static and dynamic study of mobile applications by defining “application behavior” as a possible way of understanding mobile applications and creating access control policies for them. Our user study also shows that unlike our behavior-based policy, a “deny by default” mechanism hampers usability of access control systems. We also show that inclusion of crowd-sourced policies leads to further reduction in user burden and need for engagement while capturing context-based access control policy. We enrich knowledge about mobile “application behavior” and expose this knowledge through the Mobipedia knowledge-base. We also extend context synthesis for semantic presence detection on mobile devices by combining Bluetooth, low energy beacons and Nearby Messaging services from Google.
September 5th, 2017
Cognitive Assistance for Automating the Analysis of the Federal Acquisition Regulations System
Government regulations are critical to understanding how to do business with a government entity and receive other bene?ts. However, government regulations are also notoriously long and organized in ways that can be confusing for novice users. Developing cognitive assistance tools that remove some of the burden from human users is of potential bene?t to a variety of users. The volume of data found in United States federal government regulation suggests a multiple-step approach to process the data into machine readable text, create an automated legal knowledge base capturing various facts and rules, and eventually building a legal question and answer system to acquire understanding from various regulations and provisions. Our work discussed in this paper represents our initial efforts to build a framework for Federal Acquisition Regulations System (Title 48, Code of Federal Regulations) in order to create an efficient legal knowledge base representing relationships between various legal elements, semantically similar terminologies, deontic expressions and cross-referenced legal facts and rules.
July 16th, 2017
Deep Representation of Lyrical Style and Semantics for Music Recommendation
Abhay L. Kashyap
11:00-1:00 Thursday, 20 July 2017, ITE 346
In the age of music streaming, the need for effective recommendations is important for music discovery and a personalized user experience. Collaborative filtering based recommenders suffer from popularity bias and cold-start which is commonly mitigated by content features. For music, research in content based methods have mainly been focused in the acoustic domain while lyrical content has received little attention. Lyrics contain information about a song’s topic and sentiment that cannot be easily extracted from the audio. This is especially important for lyrics-centric genres like Rap, which was the most streamed genre in 2016. The goal of this dissertation is to explore and evaluate different lyrical content features that could be useful for content, context and emotion based models for music recommendation systems.
With Rap as the primary use case, this dissertation focuses on featurizing two main aspects of lyrics; its artistic style of composition and its semantic content. For lyrical style, a suite of high level rhyme density features are extracted in addition to literary features like the use of figurative language, profanity and vocabulary strength. In contrast to these engineered features, Convolutional Neural Networks (CNN) are used to automatically learn rhyme patterns and other relevant features. For semantics, lyrics are represented using both traditional IR techniques and the more recent neural embedding methods.
These lyrical features are evaluated for artist identification and compared with artist and song similarity measures from a real-world collaborative filtering based recommendation system from Last.fm. It is shown that both rhyme and literary features serve as strong indicators to characterize artists with feature learning methods like CNNs achieving comparable results. For artist and song similarity, a strong relationship was observed between these features and the way users consume music while neural embedding methods significantly outperformed LSA. Finally, this work is accompanied by a web-application, Rapalytics.com, that is dedicated to visualizing all these lyrical features and has been featured on a number of media outlets, most notably, Vox, attn: and Metro.
Committee: Drs. Tim Finin (chair), Anupam Joshi, Tim Oates, Cynthia Matuszek and Pranam Kolari (Walmart Labs)
July 12th, 2017
Analysis of Irregular Event Sequences using Deep Learning, Reinforcement Learning, and Visualization
11:00-1:00 Thursday 13 July 2017, ITE 346, UMBC
History is nothing but a catalogued series of events organized into data. Amazon, the largest online retailer in the world, processes over 2,000 orders per minute. Orders come from customers on a recurring basis through subscriptions or as one-off spontaneous purchases, resulting in each customer exhibiting their own behavioral pattern when it comes to the way in which they place orders throughout the year. For a company such as Amazon, that generates over $130 billion of revenue each year, understanding and uncovering the hidden patterns and trends within this data is paramount in improving the efficiency of their infrastructure ranging from the management of the inventory within their warehouses, distribution of their labor force, and preparation of their online systems for the load of users. With the ever increasingly availability of big data, problems such as these are no longer limited to large corporations but are experienced across a wide range of domains and faced by analysts and researchers each and every day.
While many event analysis and time series tools have been developed for the purpose of analyzing such datasets, most approaches tend to target clean and evenly spaced data. When faced with noisy or irregular data, it has been recommended to undergo a pre-processing step of converting and transforming the data into being regular. This transformation technique arguably interferes on a fundamental level as to how the data is represented, and may irrevocably bias the way in which results are obtained. Therefore, operating on raw data, in its noisy natural form, is necessary to ensure that the insights gathered through analysis are accurate and valid.
In this dissertation novel approaches are presented for analyzing irregular event sequences using a variety of techniques ranging from deep learning, reinforcement learning, and visualization. We show how common tasks in event analysis can be performed directly on an irregular event dataset without requiring a transformation that alters the natural representation of the process that the data was captured from. The three tasks that we showcase include: (i) summarization of large event datasets, (ii) modeling the processes that create events, and (iii) predicting future events that will occur.
Committee: Drs. Tim Oates (Chair), Jesus Caban, Penny Rheingans, Jian Chen, Tim Finin
June 27th, 2017
Ph.D. Dissertation Defense
Dynamic Data Assimilation for Topic Modeling
9:00am Thursday, 29 June 2017, ITE 325b, UMBC
Understanding how a particular discipline such as climate science evolves over time has received renewed interest. By understanding this evolution, predicting the future direction of that discipline becomes more achievable. Dynamic Topic Modeling (DTM) has been applied to a number of disciplines to model topic evolution as a means to learn how a particular scientific discipline and its underlying concepts are changing. Understanding how a discipline evolves, and its internal and external influences, can be complicated by how the information retrieved over time is integrated. There are different techniques used to integrate sources of information, however, less research has been dedicated to understanding how to integrate these sources over time. The method of data assimilation is commonly used in a number of scientific disciplines to both understand and make predictions of various phenomena, using numerical models and assimilated observational data over time.
In this dissertation, I introduce a novel algorithm for scientific data assimilation, called Dynamic Data Assimilation for Topic Modeling (DDATM), which uses a new cross-domain divergence method (CDDM) and DTM. By using DDATM, observational data in the form of full-text research papers can be assimilated over time starting from an initial model. DDATM can be used as a way to integrate data from multiple sources and, due to its robustness, can exploit the assimilating observational information to better tolerate missing model information. When compared with a DTM model, the assimilated model is shown to have better performance using standard topic modeling measures, including perplexity and topic coherence. The DDATM method is suitable for prediction and results in higher likelihood for subsequent documents. DDATM is able to overcome missing information during the assimilation process when compared with a DTM model. CDDM generalizes as a method that can also bring together multiple disciplines into one cohesive model enabling the identification of related concepts and documents across disciplines and time periods. Finally, grounding the topic modeling process with an ontology improves the quality of the topics and enables a more granular understanding of concept relatedness and cross-domain influence.
The results of this dissertation are demonstrated and evaluated by applying DDATM to 30 years of reports from the Intergovernmental Panel on Climate Change (IPCC) along with more than 150,000 documents that they cite to show the evolution of the physical basis of climate change.
Committee Members: Drs. Tim Finin (co-advisor), Milton Halem (co-advisor), Anupam Joshi, Tim Oates, Cynthia Matuszek, Mark Cane, Rafael Alonso
June 16th, 2017
UMBC Data Science Graduate Programs
UMBC’s Data Science Master’s program prepares students from a wide range of disciplinary backgrounds for careers in data science. In the core courses, students will gain a thorough understanding of data science through classes that highlight machine learning, data analysis, data management, ethical and legal considerations, and more.
Students will develop an in-depth understanding of the basic computing principles behind data science, to include, but not limited to, data ingestion, curation and cleaning and the 4Vs of data science: Volume, Variety, Velocity, Veracity, as well as the implicit 5th V — Value. Through applying principles of data science to the analysis of problems within specific domains expressed through the program pathways, students will gain practical, real world industry relevant experience.
The MPS in Data Science is an industry-recognized credential and the program prepares students with the technical and management skills that they need to succeed in the workplace.
For more information and to apply online, see the Data Science MPS site.
June 15th, 2017
The topic of this month’s Data Science MD meetup is Getting Started with NLP, Sentiment Analysis and OpenNLP. The meeting will be 6:30-9:00pm, Monday, June 19 in Building 200 Room E100 at the JHU Applied Physics Laboratory. The meeting starts with networking and food and feature talks by two practitioners.
Brian Sacash (Deloitte & Touche): NLP and Sentiment Analysis
Natural Language Processing, the analysis of language, can be challenging if you don’t know where to start. Brian will walk through the Natural Language Tool Kit (NLTK), a Python library built for language analysis, and cover its core functionality. Through live coding he will demonstrate how to build a simple sentiment analysis engine from scratch.
Daniel Russ (NIH): It Takes a Village To Solve A Problem in Data Science
The talk will discuss a scientific case study in data science, computer-based occupational coding of free text job histories taken during epidemiological research studies. Beginning with a rationale for occupational coding, how the coding is performed, and how SOCcer is built on top of Apache OpenNLP. Throughout the talk, I will try to emphasize the importance of working as an interdisciplinary team.
See the meetup announcement to RSVP and get directions and more information.
May 15th, 2017
Ph.D. Dissertation Proposal
Modeling and Extracting information about Cybersecurity Events from Text
Tuesday, 16 May 2017, ITE 325, UMBC
People rely on the Internet to carry out much of the their daily activities such as banking, ordering food and socializing with their family and friends. The technology facilitates our lives, but also comes with many problems, including cybercrimes, stolen data and identity theft. With the large and increasing number of transaction done every day, the frequency of cybercrime events is also increasing. Since the number of security-related events is too high for manual review and monitoring, we need to train machines to be able to detect and gather data about potential cybersecurity threats. To support machines that can identify and understand threats, we need standard models to store the cybersecurity information and information extraction systems that can collect information to populate the models with data from text.
This dissertation will make two major contributions. The first is to extend our current cyber security ontologies with better models for relevant events, from atomic events like a login attempt, to an extended but related series of events that make up a campaign, to generalized events, such as an increase in denial-of-service attacks originating from a particular region of the world targeted at U.S. financial institutions. The second is the design and implementation of a event extraction system that can extract information about cybersecurity events from text and populated a knowledge graph using our cybersecurity event ontology. We will extend our previous work on event extraction that detected human activity events from news and discussion forums. A new set of features and learning algorithms will be introduced to improve the performance and adapt the system to cybersecurity domain. We believe that this dissertation will be useful for cybersecurity management in the future. It will quickly extract cybersecurity events from text and fill in the event ontology.
Committee: Drs. Tim Finin (chair), Anupam Joshi, Tim Oates and Karuna Joshi
May 15th, 2017
Jennifer Sleeman, Milton Halem, Tim Finin, and Mark Cane, Modeling the Evolution of Climate Change Assessment Research Using Dynamic Topic Models and Cross-Domain Divergence Maps, AAAI Spring Symposium on AI for Social Good, AAAI Press, March, 2017.
Climate change is an important social issue and the subject of much research, both to understand the history of the Earth’s changing climate and to foresee what changes to expect in the future. Approximately every five years starting in 1990 the Intergovernmental Panel on Climate Change (IPCC) publishes a set of reports that cover the current state of climate change research, how this research will impact the world, risks, and approaches to mitigate the effects of climate change. Each report supports its findings with hundreds of thousands of citations to scientific journals and reviews by governmental policy makers. Analyzing trends in the cited documents over the past 30 years provides insights into both an evolving scientific field and the climate change phenomenon itself. Presented in this paper are results of dynamic topic modeling to model the evolution of these climate change reports and their supporting research citations over a 30 year time period. Using this technique shows how the research influences the assessment reports and how trends based on these influences can affect future assessment reports. This is done by calculating cross-domain divergences between the citation domain and the assessment report domain and by clustering documents between domains. This approach could be applied to other social problems with similar structure such as disaster recovery.
March 18th, 2017
A Hands-on Introduction to TensorFlow and Machine Learning
Abhay Kashyap, UMBC ebiquity Lab
10:00-11:00am Tuesday, 28 March 2017, ITE346 ITE325b
As many of you know, TensorFlow is an open source machine learning library by Google which simplifies building and training deep neural networks that can take advantage of computers with GPUs. In this meeting, I will introduce some basic concepts of TensorFlow and machine learning in general. This will be a hands on tutorial where we will sit and code up some basic examples in TensorfFow. Specifically, we will use TensorFlow to implement linear regression, softmax classifiers and feed forward neural networks (MLP). You can find the Python notebooks here. If time permits, we will go over the implementation of the popular word2vec algorithm and introduce LSTMs to build language models.
What you need to know: Python and the basics of linear algebra and matrix operations. While it helps to know basics of machine learning, no prior knowledge will be assumed and there will be a gentle high level introduction to the algorithms we will implement.
What you need to bring: A laptop that has Python and pip installed. Having virtual environments set up on your computer is also a plus. (Warning: Windows-only users might be publicly shamed)