High-utility sequential pattern mining (HUSPM) can be applied in many applications such as retail, market basket analysis, click-stream analysis, healthcare data analysis, and bioinformatics. HUSPM algorithms discover useful information from data. However, looking at the dark side, the sensitive patterns can also be disclosed by the competitors, who use a HUSPM algorithm on the leaked data. Therefore, high-utility sequential pattern hiding (HUSPH) is used to protect the privacy information from HUSPM algorithms. This paper proposes three algorithms named High U tility S equential P attern H iding Using P ure A rray Structure (USHPA), High U tility S equential Pattern H iding Using P arallel Strategy (USHP), and High U tility S equential Pattern H iding Using R andom Distribution Strategy (USHR) for hiding high-utility sequential patterns on quantitative sequence datasets. These algorithms use a proposed data structure named P attern U tility S et for H iding (PUSH) to speed up the hiding process. We also introduce a metric called Privacy Factor to evaluate the quality of hiding results. The comparative experiments were conducted on real datasets to evaluate the performance of the proposed algorithms in terms of runtime, memory consumption, scalability, missing cost, and privacy factor. Results show that the proposed algorithms can efficiently sanitize the input datasets, and they outperform the compared algorithms for all metrics. 

Multi-core parallel algorithms for hiding high-utility sequential patterns

GPU acceleration of Levenshtein distance computation between long strings

This paper proposes a novel technique for the design of hybrid composite laminates. The method explores design space with several implicit decision trees in order to obtain the Pareto front, applying a number of manufacturing and structural considerations. The research is carried out using a parallelized breadth first search algorithm aided by dynamic programming and dynamic tree trimming; as a consequence the searching process is significantly accelerated. This novel procedure is applied to a well-known design case, where it identifies the best carbon-epoxy and glass-epoxy laminate combinations in terms of weight versus cost, and finds the Pareto front with less computational effort than alternative methods used in the past to solve the same problem. Since a full set of feasible solutions is produced with this new methodology, some important conclusions are obtained regarding hybrid laminate design criteria.

A novel technique for the design of hybrid composite laminates based on dynamic programming and dynamic tree trimming

Wavefront parallelism is a well-known technique for exploiting the concurrency of applications that execute nested loops with uniform data dependencies.
Recent research of such applications, which range from sequence alignment tools to partial differential equation solvers, has used GPUs to benefit from the massively parallel computing resources. To achieve optimal performance, tiling has been introduced as a popular solution to achieve a balanced workload. However, the use of hyperplane tiles increases the cost of synchronization and leads to poor data locality.
In this paper, we present a highly optimized implementation of the wavefront parallelism technique that harnesses the GPU architecture. A balanced workload and maximum resource utilization are achieved with an extremely low synchronization overhead. We design the kernel configuration to significantly reduce the minimum number of synchronizations required and also introduce an inter-block lock to minimize the overhead of each synchronization. In addition, shared memory is used in place of the L1 cache. The well-tailored mapping of the operations to the shared memory improves both spatial and temporal locality. We evaluate the performance of our proposed technique for four different applications: sequence alignment, edit distance, summed-area table, and 2D-SOR. The performance results demonstrate that our method achieves speedups of up to six times compared to the previous best-known hyperplane tiling-based GPU implementation.

Memory-Optimized Wavefront Parallelism on GPUs

Many software systems publish their interface using event-driven programming, where the application programmers create routines, called as responses to the events. One of such systems is the Bobox parallel frame- work, where the elements of a parallel pipeline react to events signalizing the arrival of input data. However, many algorithms are more efficiently described using classical serial approach, where the application code calls system routines to wait for required events. In this paper, we present a method of code transformation producing event- driven code from serial code and we concentrate mostly on its first part, responsible for the management of variables. Besides a friendlier programming environment, the trans- formation often improves the structure of the code with respect to compiler optimizations.

/pdf/transformation-of-pipeline-stage-algorithms-to-event-driven-5du0pnm273.pdf

Transformation of Pipeline Stage Algorithms to Event-Driven Code

Parallel and high-performance code is usually created as imperative code in FORTRAN, C, C++ with the help of parallel environments like OpenMP or Intel TBB. However, learning these languages is quite dicult com- pared to C# or Java. Although these modern languages have numerous parallel features, they lack the automatic parallelization or load distribution features known from spe- cialized parallel environments. Due to the referential nature of C# and Java, the principles of parallel environments like OpenMP cannot be directly transferred to these lan- guages. We investigated the idea of using C# as a pro- gramming language for a parallel system based on non- linear pipelines. In this paper, we propose the architecture of such system and describe some key steps that we have already taken towards the future goal of extracting both the pipeline structure and the code of the nodes from the C# source code.

/pdf/programming-parallel-pipelines-using-non-parallel-c-code-2w7sz7h9ke.pdf

Programming Parallel Pipelines Using Non-Parallel C# Code.

Modern scientific computing often combines extensive calcu- lation with complex structure of data; however, the programming method- ologies and languages of high-performance computing significantly dier from those of databases. This impedance mismatch leads many projects to the use of either primitive (like JSON) or overly general (like dis- tributed file systems) methods of data access, ignoring the decades of development in database technology. In this paper, we investigate the possibility to represent procedural code fragments using a network of op- erators similar to query plans used in relational database systems. Such a unified representation forms the necessary step towards an integrated computational-database platform. We propose a flow graph representa- tion that allows us to analyze, transform and optimize applications more eciently and without additional data. Along with the graph, we de- signed an algorithm that transforms a procedural code into the graph.

/pdf/procedural-code-representation-in-a-flow-graph-vk33l1tf22.pdf

Procedural Code Representation in a Flow Graph

Streaming environments and similar parallel platforms are widely used in image, signal, or general data processing as a means of achieving high performance. Unfortunately, they are often associated with specific programming languages and, thus, hardly accessible for non-experts. In this paper, we present a framework for transformation of a C# procedural code to a Hybrid Flow Graph – a novel intermediate code which employs the streaming paradigm and can be further converted into a streaming application. This approach will allow creating streaming applications or their parts using a widely known imperative language instead of an intricate language specific to streaming. In this paper, we focus on the transformation of control flow which represents the main difference between procedural code, driven by control flow constructs, and streaming environments, driven by data. Since the use of a streaming platform automatically enables parallelism and vectorization, we were able to demonstrate that the streaming applications generated by our method may outperform their original C# implementation.

Transforming Procedural Code for Streaming Environments

Transforming procedural code for execution by specialized parallel platforms requires a model of compu- tation sufficiently close to both the sequential program- ming languages and the target parallel environment. In this paper, we present Hybrid Flow Graphs, encompassing both control flow and data flow in a unified, pipeline based model of computation. Besides the definition of the Hy- brid Flow Graph, we introduce a formal framework based on graph rewriting, used for the specification of Hybrid Flow Graph semantics as well as for the proofs of cor- rectness of the associated code transformations. As a for- malism particularly close to pipeline-based runtime envi- ronments which include many modern database engines, the Hybrid Flow Graphs may become a powerful means for the automatic parallelization of sequential code under these environments.

/pdf/hybrid-flow-graphs-towards-the-transformation-of-sequential-14d1kctps4.pdf

Hybrid Flow Graphs: Towards the Transformation of Sequential Code into Parallel Pipeline Networks

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Michal Brabec

Author Tools

Chat about Author