scholarly journals Recurrence Networks in Natural Languages

Entropy ◽  
2019 ◽  
Vol 21 (5) ◽  
pp. 517
Author(s):  
Edgar Baeza-Blancas ◽  
Bibiana Obregón-Quintana ◽  
Candelario Hernández-Gómez ◽  
Domingo Gómez-Meléndez ◽  
Daniel Aguilar-Velázquez ◽  
...  
Keyword(s):  
Hamming Distance ◽  
Threshold Value ◽  
Network Density ◽  
Natural Languages ◽  
Linear Discriminant ◽  
Word Structure ◽  
European Languages ◽  
Written Texts ◽  
Linguistic Family ◽  
Network Metrics

We present a study of natural language using the recurrence network method. In our approach, the repetition of patterns of characters is evaluated without considering the word structure in written texts from different natural languages. Our dataset comprises 85 ebookseBooks written in 17 different European languages. The similarity between patterns of length m is determined by the Hamming distance and a value r is considered to define a matching between two patterns, i.e., a repetition is defined if the Hamming distance is equal or less than the given threshold value r. In this way, we calculate the adjacency matrix, where a connection between two nodes exists when a matching occurs. Next, the recurrence network is constructed for the texts and some representative network metrics are calculated. Our results show that average values of network density, clustering, and assortativity are larger than their corresponding shuffled versions, while for metrics like such as closeness, both original and random sequences exhibit similar values. Moreover, our calculations show similar average values for density among languages which that belong to the same linguistic family. In addition, the application of a linear discriminant analysis leads to well-separated clusters of family languages based on based on the network-density properties. Finally, we discuss our results in the context of the general characteristics of written texts.

scholarly journals Using Network Metrics in Soccer: A Macro-Analysis

2015 ◽  
Vol 45 (1) ◽  
pp. 123-134 ◽  
Author(s):  
Filipe Manuel Clemente ◽  
Micael Santos Couceiro ◽  
Fernando Manuel Lourenço Martins ◽  
Rui Sousa Mendes
Keyword(s):  
Decision Making ◽  
Network Density ◽  
Training Process ◽  
Match Analysis ◽  
Star Topology ◽  
Mean Values ◽  
Network Methods ◽  
The Mean ◽  
Network Metrics ◽  
Support Decision Making

Abstract The aim of this study was to propose a set of network methods to measure the specific properties of a team. These metrics were organised at macro-analysis levels. The interactions between teammates were collected and then processed following the analysis levels herein announced. Overall, 577 offensive plays were analysed from five matches. The network density showed an ambiguous relationship among the team, mainly during the 2nd half. The mean values of density for all matches were 0.48 in the 1st half, 0.32 in the 2nd half and 0.34 for the whole match. The heterogeneity coefficient for the overall matches rounded to 0.47 and it was also observed that this increased in all matches in the 2nd half. The centralisation values showed that there was no ‘star topology’. The results suggest that each node (i.e., each player) had nearly the same connectivity, mainly in the 1st half. Nevertheless, the values increased in the 2nd half, showing a decreasing participation of all players at the same level. Briefly, these metrics showed that it is possible to identify how players connect with each other and the kind and strength of the connections between them. In summary, it may be concluded that network metrics can be a powerful tool to help coaches understand team’s specific properties and support decision-making to improve the sports training process based on match analysis.

scholarly journals The Influence of a Network’s Spatial Symmetry, Topological Dimension, and Density on Its Percolation Threshold

Symmetry ◽  
2019 ◽  
Vol 11 (7) ◽  
pp. 920 ◽  
Author(s):  
Dmitry O. Zhukov ◽  
Elena G. Andrianova ◽  
Sergey A. Lesko
Keyword(s):  
Percolation Threshold ◽  
Mirror Symmetry ◽  
Information Transfer ◽  
3D Structure ◽  
Threshold Value ◽  
Network Density ◽  
Network Structures ◽  
Topological Dimension ◽  
Spatial Symmetry ◽  
3D Networks

Analyses of the processes of information transfer within network structures shows that the conductivity and percolation threshold of the network depend not only on its density (average number of links per node), but also on its spatial symmetry groups and topological dimension. The results presented in this paper regarding conductivity simulation in network structures show that, for regular and random 2D and 3D networks, an increase in the number of links (density) per node reduces their percolation threshold value. At the same network density, the percolation threshold value is less for 3D than for 2D networks, whatever their structure and symmetry may be. Regardless of the type of networks and their symmetry, transition from 2D to 3D structures engenders a change of percolation threshold by a value exp{−(d − 1)} that is invariant for transition between structures, for any kind of network (d being topological dimension). It is observed that in 2D or 3D networks, which can be mutually transformed by deformation without breaking and forming new links, symmetry of similarity is observed, and the networks have the same percolation threshold. The presence of symmetry axes and corresponding number of symmetry planes in which they lie affects the percolation threshold value. For transition between orders of symmetry axes, in the presence of the corresponding planes of symmetry, an invariant exists which contributes to the percolation threshold value. Inversion centers also influence the value of the percolation threshold. Moreover, the greater the number of pairs of elements of the structure which have inversion, the more they contribute to the fraction of the percolation threshold in the presence of such a center of symmetry. However, if the center of symmetry lies in the plane of mirror symmetry separating the layers of the 3D structure, the mutual presence of this group of symmetry elements do not affect the percolation threshold value. The scientific novelty of the obtained results is that for different network structures, it was shown that the percolation threshold for the blocking of nodes problem could be represented as an additive set of invariant values, that is, as an algebraic sum, the value of the members of which is stored in the transition from one structure to another. The invariant values are network density, topological dimension, and some of the elements of symmetry (axes of symmetry and the corresponding number of symmetry planes in which they lie, centers of inversion).

Research on Improved Clustering Algorithm on Web Usage Mining Based on Scientific Analysis of Web Materials

2011 ◽  
Vol 63-64 ◽  
pp. 863-867 ◽  
Author(s):  
Bin Li ◽  
Jin Yang ◽  
Cai Ming Liu ◽  
Jian Dong Zhang ◽  
Yan Zhang
Keyword(s):  
Clustering Algorithm ◽  
Hamming Distance ◽  
Clustering Algorithms ◽  
Distance Matrix ◽  
Threshold Value ◽  
Web Usage Mining ◽  
Web Usage ◽  
User Clustering ◽  
Similar Index ◽  
Browsing Behavior

Clustering analysis is an important method to research the Web user’s browsing behavior and identify the potential customers on Web usage mining. The traditional user clustering algorithms are not quite accurate. In this paper, we give two improved user clustering algorithms, which are based on the associated matrix of the user’s hits in the process of browsing website. To this matrix, an improved Hamming distance matrix is generated by defining the minimum norm or the generalized relative Hamming distance between any two vectors. Then, similar user clustering are obtained by setting the threshold value. At the last step of our algorithm, the clustering results are confirmed by defining the clustering’s Similar Index and setting sub-algorithm. Finally, the testing examples show that the new algorithms are more accurate than the old one, and the real log data presents that the improved algorithms are practical.

scholarly journals Language and the African Philosophical Traditions

2021 ◽  
Vol 2 (2) ◽  
pp. 1-19
Author(s):  
Kọ́lá Abímbọ́lá
Keyword(s):  
Cultural Practices ◽  
African Literature ◽  
African Philosophy ◽  
Philosophical Issue ◽  
Natural Languages ◽  
Conceptual Schemes ◽  
European Languages ◽  
Universal Principles ◽  
Reflective Reasoning

Are there universal principles, categories, or forms of reasoning that apply to all aspects of human experience—irrespective of culture and epoch? Numerous scholars have explored this very question from Africana perspectives: Kwasi Wiredu (1996) explored the philosophical issue of whether there are culturally defined values and concepts; Hallen and Sodipo (1986) examined the question of whether there are unique African indigenous systems of knowledge; Ngũgĩ wa Thiong’o (1994) evaluated the role of colonialism in the language of African literature; Oyerò nkẹ ́ ́ Oyěwumi (1997) argued that “gender” is a Western cultural invention that is foreign to Yorùbá systems of sociation; and Helen Veran (2001) argued that even though science, mathematics, and logic are not culturally relative, “certainty” is nonetheless derived from cultural practices and associations. Building on these and other works, this essay argues that: (i) incommensurability of “worldviews,” “perspectives,” “paradigms,” or “conceptual schemes” springs from deeper, more fundamental cognitive categories of logic that are coded into natural languages; and that (ii) consequently, as long as African reflective reasoning is expressed solely (or predominantly) in European languages, the authenticity of the “African” in African philosophy is questionable.

scholarly journals Universality Versus Language-Specificity in Listening to Running Speech

2002 ◽  
Vol 13 (3) ◽  
pp. 258-262 ◽  
Author(s):  
Anne Cutler ◽  
Katherine Demuth ◽  
James M. McQueen
Keyword(s):  
Word Recognition ◽  
Spoken Language ◽  
Word Structure ◽  
European Languages ◽  
Bantu Language ◽  
Automatic Activation

Recognizing spoken language involves automatic activation of multiple candidate words. The process of selection between candidates is made more efficient by inhibition of embedded words (like egg in beg) that leave a portion of the input stranded (here, b). Results from European languages suggest that this inhibition occurs when consonants are stranded but not when syllables are stranded. The reason why leftover syllables do not lead to inhibition could be that in principle they might themselves be words; in European languages, a syllable can be a word. In Sesotho (a Bantu language), however, a single syllable cannot be a word. We report that in Sesotho, word recognition is inhibited by stranded consonants, but stranded monosyllables produce no more difficulty than stranded bisyllables (which could be Sesotho words). This finding suggests that the viability constraint which inhibits spurious embedded word candidates is not sensitive to language-specific word structure, but is universal.

Bivalent Verb Classes in the Languages of Europe

2014 ◽  
Vol 4 (1) ◽  
pp. 116-166 ◽  
Author(s):  
Sergey Say
Keyword(s):  
Mutual Information ◽  
Hamming Distance ◽  
Pairwise Comparison ◽  
Distance Metric ◽  
Verb Classes ◽  
Low Level ◽  
European Languages ◽  
Genetic Patterns ◽  
Distance Matrices ◽  
Western European

The aims of this study are twofold: to propose methods for measuring (dis)similarities in the organization of valency class systems across languages, and to test them on a sample of European languages in order to reveal areal and genetic patterns. The data were gathered for 29 languages using a questionnaire containing 130 contextualized uses of bivalent predicates. The properties under study include (i) lexical range of transitives, (ii) lexical range of valency frames defined in terms of the “locus” of non-transitivity (whether A or P arguments are encoded by oblique devices), (iii) overall complexity of valency class systems, and (iv) lexical distribution of verbs among valency classes. In case of the simpler properties (i)–(iii), maps with quantified isoglosses and pairwise comparison of languages based on Hamming distance are used. For (iv) these methods are inapplicable (valency classes cannot be equated across languages), and I propose a distance metric based on entropy and pairwise mutual information between distributions. The distance matrices are analyzed using the NeighborNet algorithm as implemented in SplitsTree. I argue that more holistic properties of valency class systems are indicative of large areal effects: e.g., many western European languages (Germanic, Romance, Basque and some Balkan languages) are lexically “most transitive” in Europe. Low-level areal signal is clearly discernible in the data on more subtle aspects of the organization of valency classes. The findings imply that distributions of verbs into valency classes can develop quickly and are transferable in contact situations, despite drastic dissimilarities in argument-coding devices.

scholarly journals Smart Security System for Theft Protection Using Face Recognition

2021 ◽  
pp. 310-313
Author(s):  
Snehal Bhowate ◽  
Kanchan Bashine ◽  
Poonam Gajbhiye ◽  
Samiksha Paidlewar ◽  
Nikhil Dharpure ◽  
...  
Keyword(s):  
Real Time ◽  
Euclidean Distance ◽  
User Authentication ◽  
Threshold Value ◽  
Safety System ◽  
Vehicle Safety ◽  
Linear Discriminant ◽  
Parking Lot ◽  
The Face ◽  
Face Detection And Recognition

The number of car robbery attempts at the local and international scale is rising rapidly in this modern era. By inventing robbery techniques, the owners are afraid that their cars will be robbed from their ordinary parking lot or from outside. This makes vehicle protection against robbery important as a result of insecurity. The computer vision based real-time vehicle safety system solves this problem. The proposed car safety system carries through real time user authentication based on image processing using face detection and recognition techniques and a microprocessor-based control system attached to the car. The infrarot sensor attached to the driver's vehicles seat activates the hidden camera, which is fixed inside the vehicle, as the person enters the parked vehicle overcoming the existing security features. The person's face is detected using Viola Jones algorithm once the image is obtained from the activated camera. The extracted face is recognized using the improved Linéar Discriminant Analysis (LDA) algorithm that distinguishes many features rather than looking for an exact pattern based on the Euclidean distance. Authorization requires that the threshold value is established and compared to the Euclidean distance over which the person is not authenticated. The face is sent to the mobile of the owner as an MMS via the operating GSM modem, which is classified as unknown. The owner shall be controlled with the relay in accordance with the owner’s command when the information is received. The way to authenticate the person would be efficient and efficient in terms of vehicle safety.

scholarly journals Reliance on Predicative Units as a Method of Analysing and Translating Amharic Written Texts

Aethiopica ◽  
2020 ◽  
Vol 22 ◽  
Author(s):  
Iosif Fridman
Keyword(s):  
Foreign Students ◽  
The Other ◽  
Second Step ◽  
Detailed Knowledge ◽  
Syntactic Structures ◽  
The Third ◽  
European Languages ◽  
Written Texts ◽  
Verb Forms ◽  
Finite Verb

In analysing and translating Amharic texts, most foreign students have experienced major problems while trying to ‘redirect’ the rigidly leftbranching syntax of Amharic into the predominantly right-branching syntax of most European languages. The way out of this difficulty proposed by some teachers of Amharic consists in the so-called ‘translating from the end’ principle: the student begins to decipher the structure of an Amharic sentence from the finite verb form at its very end and gradually proceeds towards the beginning of the sentence, untangling—one by one—the syntactic structures involved. In the course of teaching Amharic, I have found this method largely inadequate for the purpose it is supposed to achieve. As an alternative to the ‘translating from the end’ method the author proposes another strategy which could be termed ‘reliance on predicative units’. In using this strategy, the student should, first of all, single out verb forms which are likely to perform the function of (final or dependent) predicates. The second step consists in delimiting groups, or units, headed by every such verb form. The third step is to provide a rough, working translation of every such unit without taking into consideration its relations to the other units in the sentence. The fourth, and final, step consists in joining the translations of the predicative units together; at this stage, detailed knowledge of Amharic morphosyntactic rules is very much required.

scholarly journals IRANO — SLAVICA: HISTORICAL-WORD-BUILDING PARALLELS

2019 ◽  
Vol 18 (28) ◽  
pp. 95-108
Author(s):  
Alexander Iliadi
Keyword(s):  
Conversation Analysis ◽  
Linguistic Features ◽  
Word Structure ◽  
European Languages ◽  
Language Groups ◽  
Potential Possibility ◽  
Great Relevance ◽  
The Common ◽  
General Linguistics ◽  
Selection Of

The paper deals with the findings of the research dedicated to the study of word-building in the Iranian and Slavonic languages in the comparativehistorical aspect. The task of the article was comparative analysis of Iranian and Slavonic lexemes with common Indo-European roots in diachrony and synchrony. Particularly, their etymology and peculiarities of functioning have been reviewed. In the course of the research the hypothesis of the common Indo-European legacy for word-building of the two language groups (Iranian and Slavonic) has been proved. At the same time the evidence for the common innovations for the age of Slavonic and Iranian contacts has been found. The prototypes and derivatives have been analyzed in detail with the selection of typologically common and specific linguistic features of derivation. The methodology of this research involved the inductive and deductive methods, the method of contrastive analysis and ethnic methodological conversation analysis. The analyzed word-building parallels and the conclusions are of great relevance for both comparative and historic and general linguistics. Comparativistics also employs the typological reference point. It is not only the presence of morphologically identical and chronologically similar complexes (combinations of morphemes) in two typologically not distant languages that is important. There should also be typological similarity of the processes of the word structure modification in case one and the same element is used. This proves the potential possibility of the equal development of the group of non-distantly related units in different languages. The perspective is seen in reviewing this issue in the different groups of the Indo-European languages.

scholarly journals The Spread of Ideas in a Network—The Garbage-Can Model

Entropy ◽  
2021 ◽  
Vol 23 (10) ◽  
pp. 1345
Author(s):  
Dorota Żuchowska-Skiba ◽  
Maria Stojkow ◽  
Malgorzata J. Krawczyk ◽  
Krzysztof Kułakowski
Keyword(s):  
Social Networks ◽  
Bounded Rationality ◽  
Chain Length ◽  
Hamming Distance ◽  
Cooperative Behavior ◽  
Threshold Value ◽  
Temporal Networks ◽  
Garbage Can Model ◽  
The Third ◽  
Organizational Choice

The main goal of our work is to show how ideas change in social networks. Our analysis is based on three concepts: (i) temporal networks, (ii) the Axelrod model of culture dissemination, (iii) the garbage can model of organizational choice. The use of the concept of temporal networks allows us to show the dynamics of ideas spreading processes in networks, thanks to the analysis of contacts between agents in networks. The Axelrod culture dissemination model allows us to use the importance of cooperative behavior for the dynamics of ideas disseminated in networks. In the third model decisions on solutions of problems are made as an outcome of sequences of pseudorandom numbers. The origin of this model is the Herbert Simon’s view on bounded rationality. In the Axelrod model, ideas are conveyed by strings of symbols. The outcome of the model should be the diversity of evolving ideas as dependent on the chain length, on the number of possible values of symbols and on the threshold value of Hamming distance which enables the combination.

Sign in / Sign up

Export Citation Format

Share Document