New Releases by Jeremy Kubica

Enter the wonderful world of graph algorithms, where you’ll learn when and how to apply these highly useful data structures to solve a wide range of fascinating (and fantastical) computational problems. Graph Algorithms the Fun Way offers a refreshing approach to complex concepts by blending humor, imaginative examples, and practical Python implementations to reveal the power and versatility of graph based problem-solving in the real world. Through clear diagrams, engaging examples, and Python code, you’ll build a solid foundation for addressing graph problems in your own projects. Explore a rich landscape of cleverly constructed scenarios where: Hedge mazes illuminate depth-first search Urban explorations demonstrate breadth-first search Intricate labyrinths reveal bridges and articulation points Strategic planning illustrates bipartite matching From fundamental graph structures to advanced topics, you will: Implement powerful algorithms, including Dijkstra’s, A*, and Floyd-Warshall Tackle puzzles and optimize pathfinding with newfound confidence Uncover real-world applications in social networks and transportation systems Develop robust intuition for when and why to apply specific graph techniques Delve into topological sorting, minimum spanning trees, strongly connected components, and random walks. Confront challenges like graph coloring and the traveling salesperson problem. Prepare to view the world through the lens of graphs—where connections reveal insights and algorithms unlock new possibilities.

Learn how and when to use the right data structures in any situation, strengthening your computational thinking, problem-solving, and programming skills in the process. This accessible and entertaining book provides an in-depth introduction to computational thinking through the lens of data structures — a critical component in any programming endeavor. You’ll learn how to work with more than 15 key data structures, from stacks, queues, and caches to bloom filters, skip lists, and graphs. You’ll also master linked lists by virtually standing in line at a cafe, hash tables by cataloging the history of the summer Olympics, and Quadtrees by neatly organizing your kitchen cabinets, all while becoming familiar with basic computer science concepts, like recursion and running time analysis.

Meet Frank Runtime. Disgraced ex-detective. Hard-boiled private eye. Search expert. When a robbery hits police headquarters, it’s up to Frank Runtime and his extensive search skills to catch the culprits. In this detective story, you’ll learn how to use algorithmic tools to solve the case. Runtime scours smugglers’ boats with binary search, tails spies with a search tree, escapes a prison with depth-first search, and picks locks with priority queues. Joined by know-it-all rookie Officer Notation and inept tag-along Socks, he follows a series of leads in a best-first search that unravels a deep conspiracy. Each chapter introduces a thrilling twist matched with a new algorithmic concept, ending with a technical recap. Perfect for computer science students and amateur sleuths alike, The CS Detective adds an entertaining twist to learning algorithms. Follow Frank’s mission and learn: –The algorithms behind best-first and depth-first search, iterative deepening, parallelizing, binary search, and more –Basic computational concepts like strings, arrays, stacks, and queues –How to adapt search algorithms to unusual data structures –The most efficient algorithms to use in a given situation, and when to apply common-sense heuristic methods

"The Best Practices of Spell Design introduces practical aspects of software development that are often learned through painful experience. Through Marcus and Shelly's quest, the story encourages readers to think about how to write readable, well-tested and maintainable programs."--Page 4 of cover

"Introduces principles of computational thinking, illustrating high-level computer science concepts, the motivation behind them, and their application in a non-computer fairy tale domain."--Amazon.com.

Abstract: "Discovering underlying structure from co-occurrence data is an important task in many fields, including: insurance, intelligence, criminal investigation, epidemiology, human resources, and marketing. For example a store may wish to identify underlying sets of items purchased together or a human resources department may wish to identify groups of employees that collaborate with each other. Previously Kubica et. al. presented the group detection algorithm (GDA) -- an algorithm for finding underlying groupings of entities from co-occurrence data. This algorithm is based on a probabilistic generative model and produces coherent groups that are consistent with prior knowledge. Unfortunately, the optimization used in GDA is slow, making it potentially infeasible [sic] for many real world data sets. For example, in the co-publication domain the MEDLINE database of medical publications alone contains over 2 million papers published within just a 5 year period, 1995-1999 [14]. To this end, we present k-groups -- an algorithm that uses an approach similar to that of k-means (hard clustering and localized updates) to significantly accelerate the discovery of the underlying groups while retaining GDA's probabilistic model. In addition, we show that k-groups is guaranteed to converge to a local minimum. We also compare the performance of GDA and k-groups on several real world and artificial data sets, showing the k-groups' sacrifice in solution quality is significantly offset by its increase in speed. This trade-off makes group detection tractable on significantly larger data sets."

Abstract: "Real world data is never as perfect as we would like it to be and can often suffer from corruptions that may impact interpretations of the data, models created from the data, and decisions made based on the data. One approach to this problem is outlier detection or anomaly detection in which an algorithm identifies and removes entire suspect records. But if only certain fields in a record have been corrupted then useful uncorrupted data will also be thrown out. In this paper we present an approach for identifying corrupted fields and using the remaining non-corrupted fields for subsequent modeling and analysis. Our approach learns a probabilistic model from the data that contains three components: a generative model of the clean data points, a generative model of the noise values, and a probabilistic model of the corruption process. We provide an algorithm for the unsupervised discovery of such models and empirically evaluate both its performance at detecting corrupted fields and the resulting improvement this gives to a classifier."