Pattern String Research Articles

An Effective Phishing Detection Model Based on Character Level Convolutional Neural Network from URL

Phishing is the easiest way to use cybercrime with the aim of enticing people to give accurate information such as account IDs, bank details, and passwords. This type of cyberattack is usually triggered by emails, instant messages, or phone calls. The existing anti-phishing techniques are mainly based on source code features, which require to scrape the content of web pages, and on third-party services which retard the classification process of phishing URLs. Although the machine learning techniques have lately been used to detect phishing, they require essential manual feature engineering and are not an expert at detecting emerging phishing offenses. Due to the recent rapid development of deep learning techniques, many deep learning-based methods have also been introduced to enhance the classification performance. In this paper, a fast deep learning-based solution model, which uses character-level convolutional neural network (CNN) for phishing detection based on the URL of the website, is proposed. The proposed model does not require the retrieval of target website content or the use of any third-party services. It captures information and sequential patterns of URL strings without requiring a prior knowledge about phishing, and then uses the sequential pattern features for fast classification of the actual URL. For evaluations, comparisons are provided between different traditional machine learning models and deep learning models using various feature sets such as hand-crafted, character embedding, character level TF-IDF, and character level count vectors features. According to the experiments, the proposed model achieved an accuracy of 95.02% on our dataset and an accuracy of 98.58%, 95.46%, and 95.22% on benchmark datasets which outperform the existing phishing URL models.

Infinite family of bc8 -like metastable phases in silicon

We show that new silicon crystalline phases, observed in the experiment with the laser-induced microexplosions inside silicon crystals (Rapp et al. // Nat. Commun. 6, 7555 (2015)), are all superstructures of a disordered high-symmetry phase with $Ia\bar{3}d$ cubic space group, as well as known for many years phases $bc8$ (Si-III) and $r8$ (Si-XII). The physics of this phenomenon is rather nontrivial: The $bc8$-like superstructures appear as regularly ordered patterns of switchable atomic strings, preserving everywhere the energetically favorable tetrahedral coordination of silicon atoms. The variety of superstructures arises because each string can be switched between two states independently of the others. An infinite family of different phases can be obtained this way and a number of them are considered here in detail. In addition to the known $bc8$, $bt8$, and $r8$ crystals, 128 tetrahedral structures with 16 (6 phases), 24 (22 phases), and 32 (100 phases) atoms per primitive cell are generated and studied, most of them are new ones. For the coarse-grain description of the structures with two possible states of switchable strings, the black/white (switched/nonswitched) Shubnikov symmetry groups has been used. An ab initio relaxation of the atomic positions and lattice parameters shows that all the considered phases are metastable and have higher density and energy relative to the $bc8$ phase at the ambient pressure. A possible scenario for appearance of those phases from the high-temperature amorphous phase is discussed.

Fuzzy String Matching Procedure

Background: There are different methodologies for DNA comparison based on two string algorithms, which are dependent on crisp logical principles, where there is no room for verbal (linguistic) uncertainty. These are successfully applicable procedures in DNA bioinformatics researches even by taking into consideration probabilistic random variability components based on the probability distribution functions of various types. Objective: The main purpose of this paper is to review first briefly all available DNA string matching methodologies that are based on crisp logic and then to suggest a new method based on the fuzzy logic rules and application. Methods: There are different methodologies for DNA comparison based on two string algorithms, which are dependent on crisp logical principles, where there is no room for verbal (linguistic) uncertainty. These are successfully applicable procedures in DNA bioinformatics researchers even by taking into consideration probabilistic random variability components based on the probability distribution functions of various types. Results: Fuzzy number representation of each gene implies some sort of uncertainty or unhealthiness in some or all the genes. Their better identifications can be achieved on the basis of fuzzy numbers with different membership degrees, which imply the unhealthiness or healthiness of the genes and their collective behaviors. Conclusion: After the development of fuzzy number representation of the text string coupled with crisp pattern string their relationships are searched at different shift operations, and hence, the possibility of defaulters are identified in the text string with a certain degree of membership.

On modification of Boyer-Moore-horspool's algorithm for tree pattern matching in linearised trees

Tree pattern matching on ordered trees is an important problem in Computer Science. Ordered trees can be represented as strings with additional properties via various linearisations. We present a backward tree pattern matching algorithm for ordered trees for various linear representations of trees and tree patterns. The algorithm adaptations find all occurrences of a single given tree pattern which match an input tree regardless of the chosen linearisation. The algorithms preserve the properties and advantages of standard backward string pattern matching using Boyer-Moore-Horspool's bad character shift heuristics. The number of symbol comparisons in the backward tree pattern matching can be sublinear in the size of the input tree. As in the case of the string version of Boyer-Moore-Horspool's matching algorithm, the size of the bad character shift table used by the algorithm is linear in the size of the alphabet. We compare the algorithm adaptations with the algorithm using originally chosen linear representation and with the best performing previously existing algorithms based on (non-linearised) tree pattern matching using finite tree automata or stringpath matchers. We show that the presented backward tree pattern matching algorithms outperform the non-linearising ones for single pattern matching and they perform among themselves comparably.

Genomic repeats detection using Boyer-Moore algorithm on Apache Spark Streaming

Genomic repeats, i.e., pattern searching in the string processing process to find repeated base pairs in the order of Deoxyribonucleic Acid (DNA), requires a long processing time. This research builds a big-data computational model to look for patterns in strings by modifying and implementing the Boyer-Moore algorithm on Apache Spark Streaming for human DNA sequences from the Ensemble site. Moreover, we perform some experiments on cloud computing by varying different specifications of computer clusters with involving datasets of human DNA sequences. The results obtained show that the proposed computational model on Apache Spark Streaming is faster than standalone computing and parallel computing with multicore. Therefore, it can be stated that the main contribution in this research, which is to develop a computational model for reducing the computational costs, has been achieved.

Asymptotic Analysis of the kth Subword Complexity.

Patterns within strings enable us to extract vital information regarding a string’s randomness. Understanding whether a string is random (Showing no to little repetition in patterns) or periodic (showing repetitions in patterns) are described by a value that is called the kth Subword Complexity of the character string. By definition, the kth Subword Complexity is the number of distinct substrings of length k that appear in a given string. In this paper, we evaluate the expected value and the second factorial moment (followed by a corollary on the second moment) of the kth Subword Complexity for the binary strings over memory-less sources. We first take a combinatorial approach to derive a probability generating function for the number of occurrences of patterns in strings of finite length. This enables us to have an exact expression for the two moments in terms of patterns’ auto-correlation and correlation polynomials. We then investigate the asymptotic behavior for values of . In the proof, we compare the distribution of the kth Subword Complexity of binary strings to the distribution of distinct prefixes of independent strings stored in a trie. The methodology that we use involves complex analysis, analytical poissonization and depoissonization, the Mellin transform, and saddle point analysis.

TuningMalconv: Malware Detection With Not Just Raw Bytes

Malware is one of the main security threats in this digital world. Traditional security solutions can not keep space with the emerging malwares. Currently, machine learning has bear fruit in many areas and its application in security receives more and more attention. Many different features can be used as input features, such as opcodes and byte entropy. Raw bytes of binaries can also act as the inputs of machine learning without any domain knowledge. However, the input size of raw bytes is confined. And when the binary sizes differ greatly, pure raw bytes may lack the key information to make right decisions. In this paper, we implement a detection model called TuningMalconv, which uses richer features to detect malwares. TuningMalconv is made up of two layers and the two layers are independent models. Raw bytes are the input features of the first layer. If the first layer lacks the confidence to make decisions. The second layer extracts n-grams of byte codes, PE imports, string patterns in binaries, and PE section names from the binaries. And then the second layer takes these features as inputs to make the final decisions. We use two datasets to evaluate our model, a small one and a large one. The experiment results show that TuningMalconv can achieve robust performance. It takes about 1100 seconds to detect 8213 softwares, which is practical. And the overall accuracy of TuningMalconv can reach 99.03% on our small dataset and 98.69% on the large dataset. Thus TuningMalconv can perform efficient and effective malware detection with extended features beyond raw bytes.

Apache Spark Implementations for String Patterns in DNA Sequences.

The availability of numerical data grows from 1day to another in a remarkable way. New technologies of high-throughput Next-Generation Sequencing (NGS) are producing DNA sequences. Next-Generation Sequencing describes a DNA sequencing technology which has revolutionized genomic research. In this paper, we perform some experiments using a cloud infrastructure framework, namely, Apache Spark, in some sequences derived from the National Center for Biotechnology Information (NCBI). The problems we examine are some of the most popular ones, namely, Longest Common Prefix, Longest Common Substring, and Longest Common Subsequence.

Efficient pattern matching algorithm for security and Binary Search Tree (BST) based memory system in Wireless Intrusion Detection System (WIDS)

Wireless Intrusion Detection System (WIDS) has been introduced for providing enhanced security level in WLANs, owing to the numerous and potentially devastating threats against it. WIDS are software- or hardware-based. The hardware-based approaches focus on memory efficiency in pattern matching. This paper proposes a pattern matching algorithm called All-Ready State Traversal pattern matching algorithm. This algorithm constructs the state traversal machine with 1280 bytes size, and enables users to store large sized string patterns in the pattern database. The state traversal machine facilitates the easy retrieval of these patterns through the path vector. Further, the proposed work also follows a number of basic ASCII characters with 128 bytes size; and designs the memory architecture using Binary Search Tree (BST) structure. The hardware generates the addresses of input strings. State traversal machine and bits split algorithms are used to merge together the common addresses of input strings. The merged addresses are encrypted and decrypted by Blowfish algorithm to allow the valid packets and to discard the invalid packets using WIDS. Thus, the proposed algorithm provides a significant reduction in memory usage than that of Aho–Corasick algorithm.

Accelerated preprocessing in task of searching substrings in a string

Introduction.A rapid development of the systems such as Yandex, Google, etc., has predetermined the relevance of the task of searching substrings in a string, and approaches to its solution are actively investigated. This task is used to create database management systems that support associative search. Besides, it is applicable in solving information security issues and creating antivirus programs. Algorithms of searching substring in a string are used in signature-based discovery tasks.Materials and Methods. The solution to the problem is based on the Aho-Corasick algorithm which is a typical technique of searching substrings in a string. At the same time, a new approach regarding preprocessing is employed.Research Results. The possibility of constructing the transition function and suffix references through suffix arrays and special mappings, is shown. The relationship between the prefix tree and suffix arrays was investigated, which provided the development of a fundamentally new method of constructing the transition and error functions. The results obtained enable to substantially shorten the time intervals spent on the preelection processing of a set of pattern strings when using an integer alphabet. The paper lists eight algorithms. The developed algorithms are evaluated. The results obtained are compared to the formerly known. Two theorems and eight lemmas are proved. Two examples illustrating features of the practical application of the developed preprocessing procedure are given.Discussion and Conclusions. The preprocessing procedure proposed in this paper is based on the communication between the suffix array built on the ground of a set of pattern strings and the construction of transition and error functions at the initial stages of the Aho-Corasick algorithm. This approach differs from the traditional one and requires the use of algorithms providing a suffix array in linear time. Thus, the algorithms that enable to significantly reduce the time for preprocessing of a set of pattern strings under the condition of using a certain type of alphabet in comparison to the known approach proposed in the Aho- Corasick algorithm are described. The research results presented in the paper can be used in antivirus programs that apply searching for signatures of malicious data objects in the memory of a computer system. In addition, this approach to solving the problem on searching substrings in a string will significantly speed up the operation of database management systems using associative search.

Searching Process using Bayer Moore Algorithm in Medicinal Information Media

Indonesia is one of the giant centers of biodiversity, there are many types of plants that can be used as a medicine. Based on medicinal plants collection of Bogor Botanical Garden which has the potential as medicine consists of 764 species, 465 genus and 135 families. So in search of herbs manually becomes difficult. Therefore, to facilitate the search for information about the plume of drugs, required an effective and easy to use information media by the community. In this research we used Boyer Moore algorithm to execute the searching process. This research shows the process of string matching using a Boyer-Moore algorithm in the form of visual simulation by doing string searching of a pattern string, The results show that the algorithm can be used to the searching process

Indexing weighted sequences: Neat and efficient

A weighted sequence is a sequence of probability mass functions over a finite alphabet. A weighted index is a data structure constructed for a weighted sequence and a threshold 1z that, given a string pattern, reports all positions where it occurs in the weighted sequence with probability at least 1z. We present an O(nz)-time construction of an O(nz)-sized weighted index for a weighted sequence of length n that answers queries in optimal time. The previous solution by Amir et al. (2008) required O(nz2log⁡z) time and space. Our main tools are a construction of a family of ⌊z⌋ strings that carries the information about all the strings that occur in a weighted sequence and a more straightforward solution to so-called property indexing. We present applications of our weighted index, in particular in approximate and general scenarios that were introduced by Biswas et al. (2016), and provide its implementation.

Intensification of Lempel-ZIV-Welch Algorithm

There is a necessity to reduce the consumption of exclusive resources. This is achieved using data compression. The data compression is one well known technique which can reduce the file size. A plethora of data compression algorithms are available which provides compression in various ratios. LZW is one of the powerful widely used algorithms. This paper attempts to propose and apply some enhancements to LZW, hence comes out with an efficient lossless text compression scheme that can compress a given file at better compression ratio. The paper proposes three approaches which practically enhances the original algorithm. These approaches try to gain better compression ratio. In approach1, it exploits the notion of using existing string code with odd code for a newly encounter string which is reverse of existing. In approach2 it uses a choice of code length for the current compression, so avoiding the problem of dictionary overflow. In approach3 it appends some selective set of frequently encountered string patterns. So the intensified LZW method provides better compression ratio with the inclusion of the above features.

String patterns in the doped Hubbard model.

Understanding strongly correlated quantum many-body states is one of the most difficult challenges in modern physics. For example, there remain fundamental open questions on the phase diagram of the Hubbard model, which describes strongly correlated electrons in solids. In this work, we realize the Hubbard Hamiltonian and search for specific patterns within the individual images of many realizations of strongly correlated ultracold fermions in an optical lattice. Upon doping a cold-atom antiferromagnet, we find consistency with geometric strings, entities that may explain the relationship between hole motion and spin order, in both pattern-based and conventional observables. Our results demonstrate the potential for pattern recognition to provide key insights into cold-atom quantum many-body systems.

Secure string pattern query for open data initiative

The open data initiative allows the organizations from different countries to share and integrate their data to produce value added information products and services. However, the trust relationship between any two organizations can hardly be established due to the data sensitivity and their specific responsibilities. In this paper, we propose a secure string pattern query processing framework for multiple data owners. With this framework, we design an efficient and secure indexing structure called Secure String PAttern Search tree (S2PAStree). S2PAStree allows the execution of the secure string pattern query with logarithmic time complexity. We further propose a set of secure index construction protocols under the scenario that multiple data owners share and integrate their sensitive data. Finally, we conduct comprehensive experiments on three public datasets to evaluate the efficiency and the scalability of our solutions.

XWM: a high-speed matching algorithm for large-scale URL rules in wireless surveillance applications

Large-scale high-speed URL matching is a key operation in many network security systems and surveillance applications in Wireless Sensor Networks. Classic string matching algorithms are unsuitable for large-scale URL filtering due to speed or memory consumption. This paper proposes an extend Wu-Manber algorithm (XWM) which takes advantage of the encoding characteristics of the URL greatly to improve the matching performance of the algorithm. It first adopts the pattern string window selection method to optimize Wu-Manber’s hash process, and then combines hash tables and associative containers to optimize the string comparison process. The experimental results on actual 10 million patterns show that XWM can achieve speeds that are twice as fast as traditional algorithms, especially when the shortest pattern string length is longer, it is more advantageous.

The Localization Method for Ship Network Nodes Based on High Performance Matching of Multi-pattern String TMWM

ABSTRACT Zhu, H. and Zhu, L., 2018. The localization method for ship network nodes based on high performance matching of multi-pattern string TMWM. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 776–780. Coconut Creek (Florida), ISSN 0749-0208. The pattern string matching used in the current method is single pattern matching. The matching efficiency and accuracy are low, and a large number of CPU resources are used, which reduces the efficiency and accuracy of the localization of the ship network nodes. In this paper, a high performance matching method based on multi-pattern string TMWM is proposed to preprocess the ship network nodes and make multi-pattern matching, until all the strings are terminated. Application of multi-pattern matching algorithm improves the matching efficiency, and will not take up too much CPU resources, which can increase the efficiency and accuracy of localization. The use of mobile beac...

Optimization of Spectral String Data Analysis Using a Binomial Discriminator for Weak-source Detection Decisions.

Operational health physics applications, such as radiological and nuclear monitoring and detection for homeland security or radiation protection purposes, generate time sequences of independent individual measurement data. Statistical algorithms have been developed that use the analysis of patterns in the data strings to enhance the test statistic for the decision on the absence or presence of a radiation source. These hypothesis test procedures have been applied to spectral data and have been optimized for the highest rate of correct identification of a weak Cs source at constant false positive detection rates. Optimization of correct detection decisions was investigated for various string data sequence lengths and for the regions of interest in the gamma spectrum. The highest correct source identification is achieved for string data analyses of the spectral contributions that maximize a [INCREMENT]μ/σ criterion, including energy regions around and containing the photopeak, but potentially also regions in the gamma spectrum other than those photopeak energies.

On-line weighted pattern matching

A weighted sequence is a sequence of probability distributions over an alphabet of size σ. Weighted sequences arise naturally in many applications. We study the problem of weighted pattern matching in which we are given a string pattern P of length m, a weight threshold 1z, and a weighted text X arriving on-line. We say that P occurs in X at position i if the product of probabilities of the letters of P at positions i−m+1,…,i in X is at least 1z. We first discuss how to apply a known general scheme that transforms off-line pattern matching algorithms to on-line algorithms to obtain an on-line algorithm that requires O((σ+log⁡z)log⁡m) or O(σlog2⁡m) time per arriving position; with the space requirement however being O(mmin⁡(σ,z)). Our main result is a new algorithm that processes each arriving position of X in O(z+σ) time using O(m+z) extra space.

Enhancing Test Statistics by Utilizing Data Patterns in Sequential Measurement Strings in Radiation Detection

Operational health physics applications, such as radiological and nuclear monitoring and detection for homeland security or radiation protection purposes, generate time sequences of independent individual measurement data. Statistical algorithms have been developed that utilize the analysis of patterns in data strings to enhance the test statistic for the decision on the absence or presence of a radiological source. Theoretical expectations have been verified in laboratory measurements for various lengths of data strings. Null hypothesis test performance and source detection efficacy have been shown to improve compared to the traditional method of achieving a detection decision by the comparison of a measured value to a fixed decision threshold.