Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Konstantinos Kalpakis has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Knowledge Graph Fact Prediction via Knowledge-Enriched Tensor Factorization

    by Ankur Padia, Kostantinos Kalpakis, Francis Ferraro, Tim Finin

    We present a family of novel methods for embedding knowledge graphs into real-valued tensors. These tensor-based embeddings capture the ordered relations that are typical in the knowledge graphs represented by semantic web languages like RDF. Unlike many previous models, our methods can easily use prior background knowledge provided by users or extracted automatically from existing knowledge graphs. In addition to providing more robust methods for knowledge graph embedding, we provide a provably-convergent, linear tensor factorization algorithm. We demonstrate the efficacy of our models for the task of predicting new facts across eight different knowledge graphs, achieving between 5% and 50% relative improvement over existing state-of-the-art knowledge graph embedding techniques. Our empirical evaluation shows that all of the tensor decomposition models perform well when the average degree of an entity in a graph is high, with constraint-based models doing better on graphs with a small number of highly similar relations and regularization-based models dominating for graphs with relations of varying degrees of similarity.

  • Permutation entropy analysis of vital signs data for outcome prediction of patients with severe traumatic brain injury

    by Konstantinos Kalpakis, Shiming Yang, Peter F. Hu, Colin F. Mackenzie, Lynn G. Stansbury, Deborah M. Stein, Thomas M. Scalea


  • Assessing surgical training: a utility analysis of the advanced surgical skills for exposure in trauma course

    by Stacy A. Shackelford, Evan Garofalo, Megan Holmes, Konstantinos Kalpakis, Sharon M. Henry, Colin Mackenzie, Mark W. Bowyer

  • Open Access

    Noninvasive Intracranial Pressure Monitoring Using Advanced Machine Learning Techniques

    by Peter Hu, Shiming Yang, Hegang Chen, Lynn Stansbury, Catriona Miller, Katharine Colton, Konstantinos Kalpakis, Raymond Fang, Deborah M. Stein

    This project explored the use of advanced machine learning techniques to noninvasively estimate real-time intracranial pressure (ICP) in traumatic brain injury patients from continuous electronic physiological monitoring data. We hypothesized that advanced machine learning techniques could process and analyze electronic vital signs data collected noninvasively and, when correlated with central nervous system-invasive ICP monitoring data, could provide a valid analytic platform for the noninvasive monitoring of ICP for defined periods and clinical needs. We developed a machine learning algorithm that, using noninvasive vital signs features alone, could estimate the current ICP of a patient with an accuracy of + or - 4.6 mmHg. We further developed algorithms that are capable of predicting patient future ICP with + or - 1.5 mmHg 5 minutes into the future and + or - 5 mmHg standard deviation 2 hours into the future using the continuous recordings of heart rate, systolic blood pressure, mean arterial pressure, and ICP. Realization of this analytic platform will support translation of this work into robust, field-ready clinical instrumentation that permits high-quality ICP monitoring in austere care environments not suitable for central nervous system-invasive ICP monitoring.

  • Online Recovery of Missing Values in Vital Signs Data Streams Using Low-Rank Matrix Completion

    by Shiming Yang, Konstantinos Kalpakis, Colin F. Mackenzie, Lynn G. Stansbury, Deborah M. Stein, Thomas M. Scalea, Peter F. Hu

    Continuous, automated, electronic patient vital signs data are important to physicians in evaluating traumatic brain injury (TBI) patients' physiological status and reaching timely decisions for therapeutic interventions. However, missing values in the medical data streams hinder applying many standard statistical or machine learning algorithms and result in losing some episodes of clinical importance. In this paper, we present a novel approach to filling missing values in streams of vital signs data. We construct sequences of Hankel matrices from vital signs data streams, find that these matrices exhibit low-rank, and utilize low-rank matrix completion methods from compressible sensing to fill in the missing data. We demonstrate that our approach always substantially outperforms other popular fill-in methods, like k-nearest-neighbors and expectation maximization. Further, we show that our approach recovers thousands of simulated missing data for intracranial pressure, a critical stream of measurements for guiding clinical interventions and monitoring traumatic brain injuries.


  • Fast algorithms to implement N-FINDR for hyperspectral endmember extraction

    by Wei Xiong, Chein-I Chang, Konstantinos Kalpakis

    N-FINDR suffers from several issues in its practical implementation. One is the search region which is usually the entire data space. Another related issue is its excessive computation. A third issue is the use of random initial conditions which causes inconsistency in final results that can not be reproducible. This paper develops two ways to speed up the N-FINDR in computation. One is to narrow down the search region for the N-FINDR to a feasible range, called region of interest (ROI) where data sphering/thresholding and the well-known pixel purity index (PPI) are used as a preprocessing to find a desire ROI. The other is to simplify the simplex volume computation where three methods are proposed for this purpose to reduce computational complexity of matrix determinant. In addition, in order to further reduce computational complexity two sequential N-FINDR algorithms are developed which implement the N-FINDR by finding one endmember after another in sequence so that the information provided by previously found endmembers can be used to reduce computational complexity. The conducted experiments demonstrate that while the proposed fast algorithms can greatly reduce computational complexity, their performance remains as good as the N-FINDR is and is not compromised by reduction of the search region to an ROI and simplified matrix determinant.


  • Probabilistic analysis of an enhanced partitioning algorithm for the steiner tree problem in Rd

    by Konstantinos Kalpakis, Alan T. Sherman

    The Geometric Steiner Minimum Tree problem (GSMT) is to connect at minimum cost n given points (called terminals) in d‐dimensional Euclidean space. We generalize a GSMT approximation partitioning algorithm by Komlós and Shing (KS) and analyze its performance under more relaxed conditions. These generalizations have practical applications for multilayer VLSI routing. Whereas KS assumed d = 2, our algorithm works for any dimension d ⩾ 2. Moreover, whereas their analysis assumed the rectilinear norm and a uniform distribution of input points, our analysis holds for any norm on Rd and whenever the terminals are any independent identically distributed random variables taking on values in any bounded subset of Rd. Both algorithms depend on a parameter t through which the user can trade off time for solution quality. We evaluate our algorithm in terms of its performance ratio–the ratio of the cost of the Steiner tree computed by the algorithm divided by the cost of a Steiner minimum tree. Applying a probability theorem on subadditive Euclidean functionals by Steele, we prove the following: Under the aforementioned distribution of inputs, the limit as n → ∞ of the supremum of the performance ratio of our algorithm is 1 + Ot−1/d(d‐1), almost surely. This result generalizes the corresponding 1 + O(t−1/2) bound proven by KS. Along the way, we prove a useful combinatorial lemma about d‐dimensional rectangle slicings. We prove that the worst‐case time and space complexity of our algorithm is θ(n lg (n/t) + TMST(v, v) + nTSMT(t)/t) and θ(SMST(v, v) + SSMT(t)), respectively, where v ⩽ n + 2d+1 (n/t) σ(t) is the number of vertices in the resulting Steiner tree. Here, TSMT(t) and SSMT(t) are the time and space required to solve exactly any GSMT problem of size less than t; TMST(n, m) and SMST(n, m) are the time and space required to find a minimum spanning tree of a graph with n nodes and m edges; and σ(t) is the maximum number of Steiner points for any Steiner minimum tree with t terminals. For example, for R2 and the rectilinear norm, the time is O(n lg(n/t) + n lg*n + nTSMT(t)/t) and the space is O(n lg *n + SSMT(t)). © 1994 by John Wiley & Sons, Inc.