Normalized Compression Distance Research Articles

An information theory approach to stock market liquidity.

A novel methodology is introduced to dynamically analyze the complex scaling behavior of financial data across various investment horizons. This approach comprises two steps: (a) the application of a distribution-based method for the estimation of time-varying self-similarity matrices. These matrices consist of entries that represent the scaling parameters relating pairs of distributions of price changes constructed for different temporal scales (or investment horizons); (b) the utilization of information theory, specifically the Normalized Compression Distance, to quantify the relative complexity and ascertain the similarities between pairs of self-similarity matrices. Through this methodology, distinct patterns can be identified and they may delineate the levels and the composition of market liquidity. An application to the U.S. stock index S&P500 shows the effectiveness of the proposed methodology.

Functional balance at rest of hemispheric homologs assessed via normalized compression distance

IntroductionThe formation and functioning of neural networks hinge critically on the balance between structurally homologous areas in the hemispheres. This balance, reflecting their physiological relationship, is fundamental for learning processes. In our study, we explore this functional homology in the resting state, employing a complexity measure that accounts for the temporal patterns in neurodynamics.MethodsWe used Normalized Compression Distance (NCD) to assess the similarity over time, neurodynamics, of the somatosensory areas associated with hand perception (S1). This assessment was conducted using magnetoencephalography (MEG) in conjunction with Functional Source Separation (FSS). Our primary hypothesis posited that neurodynamic similarity would be more pronounced within individual subjects than across different individuals. Additionally, we investigated whether this similarity is influenced by hemisphere or age at a population level.ResultsOur findings validate the hypothesis, indicating that NCD is a robust tool for capturing balanced functional homology between hemispheric regions. Notably, we observed a higher degree of neurodynamic similarity in the population within the left hemisphere compared to the right. Also, we found that intra-subject functional homology displayed greater variability in older individuals than in younger ones.DiscussionOur approach could be instrumental in investigating chronic neurological conditions marked by imbalances in brain activity, such as depression, addiction, fatigue, and epilepsy. It holds potential for aiding in the development of new therapeutic strategies tailored to these complex conditions, though further research is needed to fully realize this potential.

Two Approaches for Detecting Pneumonia from Chest X-ray Images

Pneumonia is an infection that inflames the air sacs in the lungs. It remains the leading cause of death in children aged <5 years. This acute respiratory infection kills over 150,000 newborns yearly. We present two approaches for detecting pneumonic lungs. Both involve chest X-ray (CXR) image classification. The first approach is based on convolutional neural networks (CNN). The second approach, proposed by us, uses the theoretical notion of Kolmogorov complexity (KC), which introduces the normalized compression distance (NCD) – a way of measuring similarities between objects of different nature, such as images. The respective algorithms are described, software implementation details are presented. Experiments were conducted to enable us to choose optimal parameter values that would facilitate accurate pneumonia detection. The two procedures showed high classification quality. This convincingly indicates they were accurate in differentiating the chest X-rays. Though a known fact, the CNN approach was confirmed to be more efficient when dealing with a larger training dataset. On the other hand, the NCD-KC technique was shown to be more efficient when handling a small number of classified images. A more sensitive and more accurate pneumonia diagnosing technique that combines the strengths of both approaches is found to be feasible.

BiComp-DTA: Drug-target binding affinity prediction through complementary biological-related and compression-based featurization approach

Drug-target binding affinity prediction plays a key role in the early stage of drug discovery. Numerous experimental and data-driven approaches have been developed for predicting drug-target binding affinity. However, experimental methods highly rely on the limited structural-related information from drug-target pairs, domain knowledge, and time-consuming assays. On the other hand, learning-based methods have shown an acceptable prediction performance. However, most of them utilize several simple and complex types of proteins and drug compounds data, ranging from the protein sequences to the topology of a graph representation of drug compounds, employing multiple deep neural networks for encoding and feature extraction, and so, leads to the computational overheads. In this study, we propose a unified measure for protein sequence encoding, named BiComp, which provides compression-based and evolutionary-related features from the protein sequences. Specifically, we employ Normalized Compression Distance and Smith-Waterman measures for capturing complementary information from the algorithmic information theory and biological domains, respectively. We utilize the proposed measure to encode the input proteins feeding a new deep neural network-based method for drug-target binding affinity prediction, named BiComp-DTA. BiComp-DTA is evaluated utilizing four benchmark datasets for drug-target binding affinity prediction. Compared to the state-of-the-art methods, which employ complex models for protein encoding and feature extraction, BiComp-DTA provides superior efficiency in terms of accuracy, runtime, and the number of trainable parameters. The latter achievement facilitates execution of BiComp-DTA on a normal desktop computer in a fast fashion. As a comparative study, we evaluate BiComp's efficiency against its components for drug-target binding affinity prediction. The results have shown superior accuracy of BiComp due to the orthogonality and complementary nature of Smith-Waterman and Normalized Compression Distance measures for protein sequences. Such a protein sequence encoding provides efficient representation with no need for multiple sources of information, deep domain knowledge, and complex neural networks.

Toward a Simulation Model Complexity Measure

Is it possible to develop a meaningful measure for the complexity of a simulation model? Algorithmic information theory provides concepts that have been applied in other areas of research for the practical measurement of object complexity. This article offers an overview of the complexity from a variety of perspectives and provides a body of knowledge with respect to the complexity of simulation models. The key terms model detail, resolution, and scope are defined. An important concept from algorithmic information theory, Kolmogorov complexity, and an application of this concept, normalized compression distance, are used to indicate the possibility of measuring changes in model detail. Additional research in this area can advance the modeling and simulation body of knowledge toward the practical application of measuring simulation model complexity. Examples show that KC and NCD measurements of simulation models can detect changes in scope and detail.

Information Complexity Ranking: A New Method of Ranking Images by Algorithmic Complexity.

Predicting how an individual will perceive the visual complexity of a piece of information is still a relatively unexplored domain, although it can be useful in many contexts such as for the design of human-computer interfaces. We propose here a new method, called Information Complexity Ranking (ICR) to rank objects from the simplest to the most complex. It takes into account both their intrinsic complexity (in the algorithmic sense) with the Kolmogorov complexity and their similarity to other objects using the work of Cilibrasi and Vitanyi on the normalized compression distance (NCD). We first validated the properties of our ranking method on a reference experiment composed of 7200 randomly generated images divided into 3 types of pictorial elements (text, digits, and colored dots). In the second step, we tested our complexity calculation on a reference dataset composed of 1400 images divided into 7 categories. We compared our results to the ground-truth values of five state-of-the-art complexity algorithms. The results show that our method achieved the best performance for some categories and outperformed the majority of the state-of-the-art algorithms for other categories. For images with many semantic elements, our method was not as efficient as some of the state-of-the-art algorithms.

Evaluating Compression-Based Phylogeny Estimation in the Presence of Incomplete Lineage Sorting.

This study assesses characteristics of the normalized compression distance (NCD) technique for building phylogenetic trees from molecular data. We examined results from a mammalian biological data set as well as a collection of simulated data with varying levels of incomplete lineage sorting. The implementation of NCD we analyze is a concatenation-based, distance-based, alignment-free, and model-free phylogeny estimation method, which takes concatenated unaligned sequence data as input and outputs a matrix of distances. We compare the NCD phylogeny estimation method with various other methods, including coalescent- and concatenation-based methods.

Zgli: A Pipeline for Clustering by Compression with Application to Patient Stratification in Spondyloarthritis.

The normalized compression distance (NCD) is a similarity measure between a pair of finite objects based on compression. Clustering methods usually use distances (e.g., Euclidean distance, Manhattan distance) to measure the similarity between objects. The NCD is yet another distance with particular characteristics that can be used to build the starting distance matrix for methods such as hierarchical clustering or K-medoids. In this work, we propose Zgli, a novel Python module that enables the user to compute the NCD between files inside a given folder. Inspired by the CompLearn Linux command line tool, this module iterates on it by providing new text file compressors, a new compression-by-column option for tabular data, such as CSV files, and an encoder for small files made up of categorical data. Our results demonstrate that compression by column can yield better results than previous methods in the literature when clustering tabular data. Additionally, the categorical encoder shows that it can augment categorical data, allowing the use of the NCD for new data types. One of the advantages is that using this new feature does not require knowledge or context of the data. Furthermore, the fact that the new proposed module is written in Python, one of the most popular programming languages for machine learning, potentiates its use by developers to tackle problems with a new approach based on compression. This pipeline was tested in clinical data and proved a promising computational strategy by providing patient stratification via clusters aiding in precision medicine.

Scalable and Fast Algorithm for Constructing Phylogenetic Trees With Application to IoT Malware Clustering

With the development of IoT devices, there is a rapid increase in new types of IoT malware and variants, causing social problems. The malware’s phylogenetic tree has been used in many studies for malware clustering or better understanding of malware evolution. However, when dealing with a large-scale malware set, conventional methods for constructing a phylogenetic tree is very time-consuming or even cannot be done in a realistic time. To solve this problem, we propose a high-speed, scalable phylogenetic tree construction algorithm with a clustering algorithm to cluster it. The proposed method involves the following steps: (1) Calculating the similarity of the specimen pairs using the normalized compression distance. (2) Creating a phylogenetic tree containing all specimens, instead of calculating the similarity of all pairs of a specimen, our algorithm only calculates a small part of the similarity matrix. (3) Dividing the phylogenetic tree into clusters by applying the minimum description length criterion. In addition, we propose a new online processing algorithm to add new malware specimens into the existing phylogenetic tree sequentially. Our goal is to reduce the computational cost of constructing the phylogenetic tree and improve the clustering accuracy of our previous research. We evaluated our method’s clustering accuracy and scalability with 65,494 IoT malware specimens. The results showed that our algorithm reduced the computation by 97.52% compared with the conventional method. Our clustering algorithm achieved accuracies of 95.5% and 99.3% for clustering family name and architecture name, respectively.

Normalized compression distance to measure cortico-muscular synchronization.

The neuronal functional connectivity is a complex and non-stationary phenomenon creating dynamic networks synchronization determining the brain states and needed to produce tasks. Here, as a measure that quantifies the synchronization between the neuronal electrical activity of two brain regions, we used the normalized compression distance (NCD), which is the length of the compressed file constituted by the concatenated two signals, normalized by the length of the two compressed files including each single signal. To test the NCD sensitivity to physiological properties, we used NCD to measure the cortico-muscular synchronization, a well-known mechanism to control movements, in 15 healthy volunteers during a weak handgrip. Independently of NCD compressor (Huffman or Lempel Ziv), we found out that the resulting measure is sensitive to the dominant-non dominant asymmetry when novelty management is required (p = 0.011; p = 0.007, respectively) and depends on the level of novelty when moving the non-dominant hand (p = 0.012; p = 0.024). Showing lower synchronization levels for less dexterous networks, NCD seems to be a measure able to enrich the estimate of functional two-node connectivity within the neuronal networks that control the body.

GPU-Based Normalized Compression Distance for Satellite Images

GPU-Based Normalized Compression Distance for Satellite Images

Fast Phylogeny of SARS-CoV-2 by Compression.

The compression method to assess similarity, in the sense of having a small normalized compression distance (NCD), was developed based on algorithmic information theory to quantify the similarity in files ranging from words and languages to genomes and music pieces. It has been validated on objects from different domains always using essentially the same software. We analyze the whole-genome phylogeny and taxonomy of the SARS-CoV-2 virus, which is responsible for causing the COVID-19 disease, using the alignment-free compression method to assess similarity. We compare the SARS-CoV-2 virus with a database of over 6500 viruses. The results suggest that the SARS-CoV-2 virus is closest in that database to the RaTG13 virus and rather close to the bat SARS-like coronaviruses bat-SL-CoVZXC21 and bat-SL-CoVZC45. Over 6500 viruses are identified (given by their registration code) with larger NCDs. The NCDs are compared with the NCDs between the mtDNA of familiar species. We address the question of whether pangolins are involved in the SARS-CoV-2 virus. The compression method is simpler and possibly faster than any other whole-genome method, which makes it the ideal tool to explore phylogeny. Here, we use it for the complex case of determining this similarity between the COVID-19 virus, SARS-CoV-2 and many other viruses. The resulting phylogeny and taxonomy closely resemble earlier results from by alignment-based methods and a machine-learning method, providing the most compelling evidence to date for the compression method, showing that one can achieve equivalent results both simply and quickly.

Enhancing Reusability: An Integrated Framework for Software Requirements Classification and Prioritization

Requirements Prioritization (RP) is the process of planning, identifying, rating and organizing requirements to avoid breach of trust, contract or agreement. The need of RP is inevitable if the concerns, cost and budget estimation of the stakeholders are considered as primary parameters in a software project. In order to ensure the reusability of existing modules, the new requirements must be classified and coordinated with the current ones in the repository, which guarantees the monetary use of assets. In this research, we have suggested a new framework for software requirements classification and prioritization. This framework is called the Software Requirements Classification and Prioritization Framework (SCPF). It will help the software project managers in managing the requirements at the initial stage of software development. We have identified the key factors and developed a taxonomy in order to classify these factors. This study likewise devised a new RP algorithm, which is based on the normalized aggregation values of the identified factors, and also contributes for the requirements classification by using [Formula: see text]-Nearest-Neighbor (KNN) classifier based on the well-known Normalized Compression Distance (NCD) similarity measure. In addition to this, we have implemented the SCPF by developing a Web-based application which provides the computation features for requirements classification and prioritization. Furthermore, we have conducted a case study and an industry survey to evaluate the results of the SCPF. A comparative study is also presented to show the effectiveness of SCPF with respect to three different state-of-the-art RP approaches. The computed SCPF results are very promising and showed significant improvement in the quality of requirements classification and prioritization.

PhishSim: Aiding Phishing Website Detection With a Feature-Free Tool

In this paper, we propose a feature-free method for detecting phishing websites using the Normalized Compression Distance (NCD), a parameter-free similarity measure which computes the similarity of two websites by compressing them, thus eliminating the need to perform any feature extraction. It also removes any dependence on a specific set of website features. This method examines the HTML of webpages and computes their similarity with known phishing websites, in order to classify them. We use the Furthest Point First algorithm to perform phishing prototype extractions, in order to select instances that are representative of a cluster of phishing webpages. We also introduce the use of an incremental learning algorithm as a framework for continuous and adaptive detection without extracting new features when concept drift occurs. On a large dataset, our proposed method significantly outperforms previous methods in detecting phishing websites, with an AUC score of 98.68%, a high true positive rate (TPR) of around 90%, while maintaining a low false positive rate (FPR) of 0.58%. Our approach uses prototypes, eliminating the need to retain long term data in the future, and is feasible to deploy in real systems with a processing time of roughly 0.3 seconds.

Analysis of relaminarizing turbulent boundary layer under a favorable pressure gradient with machine learning

Relaminarization process of an accelerated turbulent boundary layer was analyzed using neural network. The neural network model was constructed and trained from the streamwise fluctuating velocity data with the aid of a hot-wire anemometer. Keras, a deep learning library in python, was used to build the model. Two types of data were used for the training data: turbulent and laminar. From the created model, the predicted probability of turbulent or laminar flow was obtained for the streamwise and wall-normal fluctuating velocities in the whole region within the relaminarization process. From the predicted probabilities, the relaminarization process was examined. The predicted probability of turbulence for both the streamwise and wall-normal fluctuating velocities decreased downstream. During the relaminarization process, range of the predicted probability of turbulence differed between streamwise and wall-normal components. The difference in directional components between the training and the test data cause a difference in range and ambiguity of the predicted probability. The predicted probability was able to extract the characteristics of the turbulence within the minimum eddy. The distribution of the predicted probability of turbulence for the streamwise fluctuating velocity differed from that of Normalized compression distance of streamwise fluctuating velocity whose reference position was within turbulent region.

Structured patterns of activity in pulse-coupled oscillator networks with varied connectivity.

Identifying coordinated activity within complex systems is essential to linking their structure and function. We study collective activity in networks of pulse-coupled oscillators that have variable network connectivity and integrate-and-fire dynamics. Starting from random initial conditions, we see the emergence of three broad classes of behaviors that differ in their collective spiking statistics. In the first class (“temporally-irregular”), all nodes have variable inter-spike intervals, and the resulting firing patterns are irregular. In the second (“temporally-regular”), the network generates a coherent, repeating pattern of activity in which all nodes fire with the same constant inter-spike interval. In the third (“chimeric”), subgroups of coherently-firing nodes coexist with temporally-irregular nodes. Chimera states have previously been observed in networks of oscillators; here, we find that the notions of temporally-regular and chimeric states encompass a much richer set of dynamical patterns than has yet been described. We also find that degree heterogeneity and connection density have a strong effect on the resulting state: in binomial random networks, high degree variance and intermediate connection density tend to produce temporally-irregular dynamics, while low degree variance and high connection density tend to produce temporally-regular dynamics. Chimera states arise with more frequency in networks with intermediate degree variance and either high or low connection densities. Finally, we demonstrate that a normalized compression distance, computed via the Lempel-Ziv complexity of nodal spike trains, can be used to distinguish these three classes of behavior even when the phase relationship between nodes is arbitrary.

A Compression-Based Method for Detecting Anomalies in Textual Data.

Nowadays, information and communications technology systems are fundamental assets of our social and economical model, and thus they should be properly protected against the malicious activity of cybercriminals. Defence mechanisms are generally articulated around tools that trace and store information in several ways, the simplest one being the generation of plain text files coined as security logs. Such log files are usually inspected, in a semi-automatic way, by security analysts to detect events that may affect system integrity, confidentiality and availability. On this basis, we propose a parameter-free method to detect security incidents from structured text regardless its nature. We use the Normalized Compression Distance to obtain a set of features that can be used by a Support Vector Machine to classify events from a heterogeneous cybersecurity environment. In particular, we explore and validate the application of our method in four different cybersecurity domains: HTTP anomaly identification, spam detection, Domain Generation Algorithms tracking and sentiment analysis. The results obtained show the validity and flexibility of our approach in different security scenarios with a low configuration burden.

Cryptographic Software IP Protection without Compromising Performance or Timing Side-channel Leakage

Program obfuscation is a widely used cryptographic software intellectual property (IP) protection technique against reverse engineering attacks in embedded systems. However, very few works have studied the impact of combining various obfuscation techniques on the obscurity (difficulty of reverse engineering) and performance (execution time) of obfuscated programs. In this article, we propose a Genetic Algorithm (GA)-based framework that not only optimizes obscurity and performance of obfuscated cryptographic programs, but it also ensures very low timing side-channel leakage. Our proposed T iming S ide C hannel S ensitive P rogram O bfuscation O ptimization F ramework (TSC-SPOOF) determines the combination of obfuscation transformation functions that produce optimized obfuscated programs with preferred optimization parameters. In particular, TSC-SPOOF employs normalized compression distance (NCD) and channel capacity to measure obscurity and timing side-channel leakage, respectively. We also use RISC-V rocket core running on a Xilinx Zynq FPGA device as part of our framework to obtain realistic results. The experimental results clearly show that our proposed solution leads to cryptographic programs with lower execution time, higher obscurity, and lower timing side-channel leakage than unguided obfuscation.

ANALISIS PENERAPAN NORMALIZED WEB DISTANCE: TINJAUAN KASUS GOOGLE DAN COMPRESSION DISTANCE

Currently, various techniques for measuring the proximity between two objects in the internet network are continuously being developed. The objects in question are in the form of concepts, e-mails, words, and so on. Normalized Web Distance (NWD) has proven to be a simple, yet powerful measure of the semantic linkages between the two concepts. NWD has several approaches according to the object being measured, such as Normalized Google Distance (NGD) and Normalized Compression Distance (NCD). NGD and NCD have a way of determining similarity and calculating the distance to find the similarity of two different measuring objects. This paper shows and provides information on the performance of the two NWD approaches, namely NGD and NCD to facilitate understanding of the use of NGD and NCD on various problems. Correct understanding can put the NGD and the NCD in the right case.

Computational analysis of the SARS-CoV-2 and other viruses based on the Kolmogorov's complexity and Shannon's information theories.

This paper tackles the information of 133 RNA viruses available in public databases under the light of several mathematical and computational tools. First, the formal concepts of distance metrics, Kolmogorov complexity and Shannon information are recalled. Second, the computational tools available presently for tackling and visualizing patterns embedded in datasets, such as the hierarchical clustering and the multidimensional scaling, are discussed. The synergies of the common application of the mathematical and computational resources are then used for exploring the RNA data, cross-evaluating the normalized compression distance, entropy and Jensen–Shannon divergence, versus representations in two and three dimensions. The results of these different perspectives give extra light in what concerns the relations between the distinct RNA viruses.