Search for: All records

The Twitter-Based Knowledge Graph for Researchers project is an effort to construct a knowledge graph of computation-based tasks and corresponding outputs. It will be utilized by subject matter experts, statisticians, and developers. A knowledge graph is a directed graph of knowledge accumulated from a variety of sources. For our application, Subject Matter Experts (SMEs) are experts in their respective non-computer science fields, but are not necessarily experienced with running heavy computation on datasets. As a result, they find it difficult to generate workflows for their projects involving Twitter data and advanced analysis. Workflow management systems and libraries that facilitate computation are only practical when the users of these systems understand what analysis they need to perform. Our goal is to bridge this gap in understanding. Our queryable knowledge graph will generate a visual workflow for these experts and researchers to achieve their project goals. After meeting with our client, we established two primary deliverables. First, we needed to create an ontology of all Twitter-related information that an SME might want to answer. Secondly, we needed to build a knowledge graph based on this ontology and produce a set of APIs to trigger a set of network algorithms based on the information queried to the graph. An ontology is simply the class structure/schema for the graph. Throughout future meetings, we established some more specific additional requirements. Most importantly, the client stressed that users should be able to bring their own data and add it to our knowledge graph. As more research is completed and new technologies are released, it will be important to be able to edit and add to the knowledge graph. Next, we must be able to provide metrics about the data itself. These metrics will be useful for both our own work, and future research surrounding graph search problems and search optimization. Additionally, our system should provide users with information regarding the original domain that the algorithms and workflows were run against. That way they can choose the best workflow for their data. The project team first conducted a literature review, reading reports from the CS5604 Information Retrieval courses in 2016 and 2017 to extract information related to Twitter data and algorithms. This information was used to construct our raw ontology in Google Sheets, which contained a set of dataset-algorithm-dataset tuples. The raw ontology was then converted into nodes and edges csv files for building the knowledge graph. After implementing our original solution on a CentOS virtual machine hosted by the Virginia Tech Department of Computer Science, we transitioned our solution to Grakn, an open-source knowledge graph database that supports hypergraph functionality. When finalizing our workflow paths, we noted some nodes depended on completion of two or more inputs, representing an ”AND” edge. This phenomenon is modeled as a hyperedge with Grakn, initiating our transition from Neo4J to Grakn. Currently, our system supports queries through the console, where a user can type a Graql statement to retrieve information about data in the graph, from relationships to entities to derived rules. The user can also interact with the data via Grakn's data visualizer: Workbase. The user can enter Graql queries to visualize connections within the knowledge graph. 

In the United States, every state has a tourism website. These sites highlight the main attractions of the state, travel tips, and blog posts among other relevant information. The funding for these websites often comes from occupancy taxes, a form of taxes that comes from tourists who stay in hotels and visit attractions. Therefore, current and past tourists fund the efforts to draw future tourists into the state. Since state tourism is funded by the success of past tourism efforts, it is important for researchers to spend their time and resources on finding out what efforts were successful and which weren’t. With this comes the importance of seeing trends in past tourism endeavors. By examining past tourism websites, patterns can be drawn about information that changed, from season to season and year to year. These patterns can be used to see what researchers deemed as successful tourism efforts, and help guide future state tourism decisions. Our client, Dr. Florian Zach of the Howard Feiertag Department of Hospitality and Tourism Management, wants to use this historical analysis on state tourism information to help with his research on trends in state tourism website content. Iterations of the California state tourism website, among other sites, are stored as snapshots on the Internet Archive and can be accessed to see changes in websites over time. Our team was given Parquet files of these snapshots dating back to 2008. The goal of the project was to assist Dr. Zach by using the California state tourism website, visitcalifornia.com, and these snapshots as an avenue to explore data extraction and visualization techniques on tourism patterns to later be expanded to other states’ tourism websites. Python’s Pandas library was utilized to examine and extract relevant pieces of data from the given Parquet files. Once the data was extracted, we used Python’s Natural Language Processing Toolkit to remove non-English words, punctuation, and a set of unimportant “stop words”. With this refined data, we were able to make visualizations regarding the frequency of words in the headers and body of the website snapshots. The data was examined in its entirety as well as in groups of seasons and years. Microsoft Excel functions were utilized to examine and visualize the data in these formats. These data extraction and visualization techniques that we became familiar with will be passed down to a future team. The research on state tourism site information can be expanded to different metadata sets and to other states. 

The purpose of the Twitter Disaster Behavior project is to identify patterns in online behavior during natural disasters by analyzing Twitter data. The main goal is to better understand the needs of a community during and after a disaster, to aid in recovery. The datasets analyzed were collections of tweets about Hurricane Maria, and recent earthquake events, in Puerto Rico. All tweets pertaining to Hurricane Maria are from the timeframe of September 15 through October 14, 2017. Similarly, tweets pertaining to the Puerto Rico earthquake from January 7 through February 6, 2020 were collected. These tweets were then analyzed for their content, number of retweets, and the geotag associated with the author of the tweet. We counted the occurrence of key words in topics relating to preparation, response, impact, and recovery. This data was then graphed using Python and Matplotlib. Additionally, using a Twitter crawler, we extracted a large dataset of tweets by users that used geotags. These geotags are used to examine location changes among the users before, during, and after each natural disaster. Finally, after performing these analyses, we developed easy to understand visuals and compiled these figures into a poster. Using these figures and graphs, we compared the two datasets in order to identify any significant differences in behavior and response. The main differences we noticed stemmed from two key reasons: hurricanes can be predicted whereas earthquakes cannot, and hurricanes are usually an isolated event whereas earthquakes are followed by aftershocks. Thus, the Hurricane Maria dataset experienced the highest amount of tweet activity at the beginning of the event and the Puerto Rico earthquake dataset experienced peaks in tweet activity throughout the entire period, usually corresponding to aftershock occurrences. We studied these differences, as well as other important trends we identified. 

Searching for documents with semantically similar content is a fundamental problem in the information retrieval domain with various challenges, primarily, in terms of efficiency and effectiveness. Despite the promise of modeling structured dependencies in documents, several existing text hashing methods lack an efficient mechanism to incorporate such vital information. Additionally, the desired characteristics of an ideal hash function, such as robustness to noise, low quantization error and bit balance/uncorrelation, are not effectively learned with existing methods. This is because of the requirement to either tune additional hyper-parameters or optimize these heuristically and explicitly constructed cost functions. In this paper, we propose a Denoising Adversarial Binary Autoencoder (DABA) model which presents a novel representation learning framework that captures structured representation of text documents in the learned hash function. Also, adversarial training provides an alternative direction to implicitly learn a hash function that captures all the desired characteristics of an ideal hash function. Essentially, DABA adopts a novel single-optimization adversarial training procedure that minimizes the Wasserstein distance in its primal domain to regularize the encoder’s output of either a recurrent neural network or a convolutional autoencoder.We empirically demonstrate the effectiveness of our proposed method in capturing the intrinsic semantic manifold of the related documents. The proposed method outperforms the current state-of-the-art shallow and deep unsupervised hashing methods for the document retrieval task on several prominent document collections. 

Electronic medical records (EMR) contain comprehensive patient information and are typically stored in a relational database with multiple tables. Effective and efficient patient information retrieval from EMR data is a challenging task for medical experts. Question-to-SQL generation methods tackle this problem by first predicting the SQL query for a given question about a database, and then, executing the query on the database. However, most of the existing approaches have not been adapted to the healthcare domain due to a lack of healthcare Question-to-SQL dataset for learning models specific to this domain. In addition, wide use of the abbreviation of terminologies and possible typos in questions introduce additional challenges for accurately generating the corresponding SQL queries. In this paper, we tackle these challenges by developing a deep learning based TRanslate-Edit Model for Question-to-SQL (TREQS) generation, which adapts the widely used sequence-to-sequence model to directly generate the SQL query for a given question, and further performs the required edits using an attentive-copying mechanism and task-specific look-up tables. Based on the widely used publicly available electronic medical database, we create a new large-scale Question-SQL pair dataset, named MIMICSQL, in order to perform the Question-to-SQL generation task in healthcare domain. An extensive set of experiments are conducted to evaluate the performance of our proposed model on MIMICSQL. Both quantitative and qualitative experimental results indicate the flexibility and efficiency of our proposed method in predicting condition values and its robustness to random questions with abbreviations and typos. 

Abstract Natural language processing (NLP) covers a large number of topics and tasks related to data and information management, leading to a complex and challenging teaching process. Meanwhile, problem-based learning is a teaching technique specifically designed to motivate students to learn efficiently, work collaboratively, and communicate effectively. With this aim, we developed a problem-based learning course for both undergraduate and graduate students to teach NLP. We provided student teams with big data sets, basic guidelines, cloud computing resources, and other aids to help different teams in summarizing two types of big collections: Web pages related to events, and electronic theses and dissertations (ETDs). Student teams then deployed different libraries, tools, methods, and algorithms to solve the task of big data text summarization. Summarization is an ideal problem to address learning NLP since it involves all levels of linguistics, as well as many of the tools and techniques used by NLP practitioners. The evaluation results showed that all teams generated coherent and readable summaries. Many summaries were of high quality and accurately described their corresponding events or ETD chapters, and the teams produced them along with NLP pipelines in a single semester. Further, both undergraduate and graduate students gave statistically significant positive feedback, relative to other courses in the Department of Computer Science. Accordingly, we encourage educators in the data and information management field to use our approach or similar methods in their teaching and hope that other researchers will also use our data sets and synergistic solutions to approach the new and challenging tasks we addressed. 

Entity linking is the task of linking mentions of named entities in natural language text, to entities in a curated knowledge-base. This is of significant importance in the biomedical domain, where it could be used to semantically annotate a large volume of clinical records and biomedical literature, to standardized concepts described in an ontology such as Unified Medical Language System (UMLS). We observe that with precise type information, entity disambiguation becomes a straightforward task. However, fine-grained type information is usually not available in biomedical domain. Thus, we propose LATTE, a LATent Type Entity Linking model, that improves entity linking by modeling the latent fine-grained type information about mentions and entities. Unlike previous methods that perform entity linking directly between the mentions and the entities, LATTE jointly does entity disambiguation, and latent fine-grained type learning, without direct supervision. We evaluate our model on two biomedical datasets: MedMentions, a large scale public dataset annotated with UMLS concepts, and a de-identified corpus of dictated doctor’s notes that has been annotated with ICD concepts. Extensive experimental evaluation shows our model achieves significant performance improvements over several state-of-the-art techniques. 

Survival analysis aims at predicting time to event of interest along with its probability on longitudinal data. It is commonly used to make predictions for a single specific event of interest at a given time point. However, predicting the occurrence of multiple events simultaneously and dynamically is needed in many applications. An intuitive way to solve this problem is to simply apply the regular survival analysis method independently to each task at each time point. However, it often leads to a suboptimal solution since the underlying dependencies between tasks are ignored, which motivates us to analyze these tasks jointly to select common features shared across all tasks. In this paper, we formulate a temporal Multi-Task learning framework (MTMT) using tensor representation. More specifically, given a survival dataset and a sequence of time points, which are considered as the monitored time points, we model each task at each time point as a regular survival analysis problem and optimize them simultaneously. We demonstrate the performance of MTMT model on two real-world datasets. We show the superior performance of the MTMT model compared to several state-of-the-art models. We also provide the list of important features selected to demonstrate the interpretability of our model. 

In the era of big data, online doctor review platforms, which enable patients to give feedback to their doctors, have become one of the most important components in healthcare systems. On one hand, they help patients to choose their doctors based on the experience of others. On the other hand, they help doctors to improve the quality of their service. Moreover, they provide important sources for us to discover common concerns of patients and existing problems in clinics, which potentially improve current healthcare systems. In this paper, we systematically investigate the dataset from one of such review platform, namely, ratemds.com, where each review for a doctor comes with an overall rating and ratings of four different aspects. A comprehensive statistical analysis is conducted first for reviews, ratings, and doctors. Then, we explore the content of reviews by extracting latent topics related to different aspects with unsupervised topic modeling techniques. As the core component of this paper, we propose a multi-task learning framework for the document-level multi-aspect sentiment classification. This task helps us to not only recover missing aspect-level ratings and detect inconsistent rating scores but also identify aspect-keywords for a given review based on ratings. The proposed model takes both features of doctors and aspect-keywords into consideration. Extensive experiments have been conducted on two subsets of ratemds dataset to demonstrate the effectiveness of the proposed model.