Proceedings of International Conference On Recent Trends in Machine Learning, Iot, Smart Cities and Applications: Icmisc 2020 Vinit Kumar Gunjan
Proceedings of International Conference On Recent Trends in Machine Learning, Iot, Smart Cities and Applications: Icmisc 2020 Vinit Kumar Gunjan
Proceedings of International Conference On Recent Trends in Machine Learning, Iot, Smart Cities and Applications: Icmisc 2020 Vinit Kumar Gunjan
com
https://textbookfull.com/product/biota-grow-2c-gather-2c-cook-loucas/
textbookfull.com
https://textbookfull.com/product/stories-without-borders-the-berlin-
wall-and-the-making-of-a-global-iconic-event-1st-edition-sonnevend/
textbookfull.com
Advances in Human Factors in Architecture Sustainable
Urban Planning and Infrastructure Proceedings of the AHFE
2020 Virtual Conference on Human Factors in Architecture
Sustainable Urban Planning and Infrastructure 16 20 July
https://textbookfull.com/product/advances-in-human-factors-in-
2020 USA Jerzy Charytonowicz
architecture-sustainable-urban-planning-and-infrastructure-
proceedings-of-the-ahfe-2020-virtual-conference-on-human-factors-in-
architecture-sustainable-urban-planning-and/
textbookfull.com
https://textbookfull.com/product/mobility-memory-and-the-lifecourse-
in-twentieth-century-literature-and-culture-lynne-pearce/
textbookfull.com
https://textbookfull.com/product/introduction-to-the-theory-of-the-
early-universe-hot-big-bang-theory-2nd-edition-gorbunov-rubakov/
textbookfull.com
https://textbookfull.com/product/hang-time-my-life-in-basketball-1st-
edition-elgin-baylor/
textbookfull.com
Advances in Intelligent Systems and Computing 1245
Proceedings
of International
Conference on Recent
Trends in Machine
Learning, IoT,
Smart Cities and
Applications
ICMISC 2020
Advances in Intelligent Systems and Computing
Volume 1245
Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
Advisory Editors
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing,
Universidad Central de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, School of Computer Science and Electronic Engineering,
University of Essex, Colchester, UK
László T. Kóczy, Department of Automation, Széchenyi István University,
Gyor, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas
at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao
Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information Technology,
University of Technology Sydney, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute
of Technology, Tijuana, Mexico
Nadia Nedjah, Department of Electronics Engineering, University of Rio de Janeiro,
Rio de Janeiro, Brazil
Ngoc Thanh Nguyen , Faculty of Computer Science and Management,
Wrocław University of Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering,
The Chinese University of Hong Kong, Shatin, Hong Kong
The series “Advances in Intelligent Systems and Computing” contains publications
on theory, applications, and design methods of Intelligent Systems and Intelligent
Computing. Virtually all disciplines such as engineering, natural sciences, computer
and information science, ICT, economics, business, e-commerce, environment,
healthcare, life science are covered. The list of topics spans all the areas of modern
intelligent systems and computing such as: computational intelligence, soft comput-
ing including neural networks, fuzzy systems, evolutionary computing and the fusion
of these paradigms, social intelligence, ambient intelligence, computational neuro-
science, artificial life, virtual worlds and society, cognitive science and systems,
Perception and Vision, DNA and immune based systems, self-organizing and
adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics
including human-machine teaming, knowledge-based paradigms, learning para-
digms, machine ethics, intelligent data analysis, knowledge management, intelligent
agents, intelligent decision making and support, intelligent network security, trust
management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are
primarily proceedings of important conferences, symposia and congresses. They
cover significant recent developments in the field, both of a foundational and
applicable character. An important characteristic feature of the series is the short
publication time and world-wide distribution. This permits a rapid and broad
dissemination of research results.
** Indexing: The books of this series are submitted to ISI Proceedings,
EI-Compendex, DBLP, SCOPUS, Google Scholar and Springerlink **
Editors
Proceedings of International
Conference on Recent Trends
in Machine Learning, IoT,
Smart Cities and Applications
ICMISC 2020
123
Editors
Vinit Kumar Gunjan Jacek M. Zurada
Department of Computer Science Department of Electrical
and Engineering and Computer Engineering
CMR Institute of Technology University of Louisville
Hyderabad, India Louisville, KY, USA
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface
Artificial intelligence (AI) and machine learning (ML) are the emerging technolo-
gies that are moving organizations faster than ever before. In this era of digital
transformation, the success is based on using analytics to find huge amount of data
with close insights. Historically, these insights were discovered manually through
in-depth data analytics–but data complexity continues to increase, as does the
complexity of data. AI and ML are the latest tools for data scientists, enabling them
to rapidly refine the data to ensure its value.
The term ‘smart cities’ first gained traction back in the 1990s, when it was
adopted as a way to illustrate the use of technology and innovation in urban
development. Since then, the world has become increasingly urbanized. By 2020,
55% of the world’s population lives in urban areas, and the number is expected to
increase to around 70% by 2050. This rapid urbanization will put increasing
pressure on resources and address the demand for intelligent and sustainable
environments. The requirements of citizens providing a higher quality of life will
increase.
We are entering into a new era of computing technology that many are calling as
the Internet of Things (IoT). Machine to Machine, Machine to Infrastructure,
Machine to Environment, Internet of Everything, Internet of Intelligent Things,
Intelligent System–call it what you want, but it is happening, and its potential is
huge.
This book is comprised of selected and presented papers of the International
conference on Recent Trends in Machine Learning, IOT, Smart Cities &
Applications, 2020. It consists of selected manuscripts arranged on the basis of their
approaches and contributions to the scope of the conference. The chapters of this
book present key algorithms and theories that form the core of the technologies and
applications concerned, consisting mainly of face recognition, evolutionary algo-
rithms such as genetic algorithms, automotive applications, automation devices
with artificial neural networks, business management systems, IoT, machine
learning, data science and modern speech processing systems. This book also
v
vi Preface
covers recent advances in medical diagnostic systems, sensor networks and systems
of VLSI domain. Discussion of learning and software modules in deep learning
algorithms is added wherever suitable. In a nut shell the book will bring insights to
the modern advancements involved in smart cities with an IoT and ML approach.
vii
viii Contents
Genetic System for Project Support with the Sequencing Problem . . . . 977
Amelec Viloria, Noel Varela, Carlos Herazo-Beltran,
Omar Bonerge Pineda Lezama, Alberto Mercado, Jairo Martinez Ventura,
and Hugo Hernandez Palma
Method for the Recovery of Images in Databases of Skin Cancer . . . . . 985
Amelec Viloria, Noel Varela, Narledys Nuñez-Bravo,
and Omar Bonerge Pineda Lezama
xvii
Automated Identification of Interictal
Activity from EEG Signal Using
Non-linear Features
1 Introduction
Set C and Set D hold interictal discharges from five patients from the epileptic zone
in set C and opposite to epileptic zone in set D. Set E contains ictal activity. In this
work, our focus is to distinguish controlled activity from interictal activity. For this
work, we have considered four cases under group controlled and interictal; which
are A–C, B–C, A–D, and B–D (Fig. 2).
2 Methodology
For this work the objective was to classify between data taken from controlled patients
and data taken from people with interictal activity; therefore, a total of four cases
were considered. Total of 100 files was present in each set. Each signal was recorded
for 23.6 s at a sampling frequency of 173.6 Hz and hence had 4096 data points.
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
4 A. Kaur et al.
All the datasets considered in this work are divided into five sub-bands; ‘db4’ wavelet
is used for the sub-band division. Table 1 depicts the specific frequency range.
Feature Extraction:
For this work, three entropy-based features are Approximate Entropy, Sample
Entropy, and fuzzy approximate entropy are extracted out of the five sub-bands used
mentioned in Table 1. Correlation dimension is extracted out of complete signal. Total
16 features are used to train the model. These entropy-based features have been also
used by previous researches either independently or with combination with other
features and have been elaborately explained in [11–15]. For this work, the value of
parameters used are as follows N is taken to be 4097, m is embedding dimension; it
is set to two and r is vector comparison distance. It is set to 0.2 times the standard
deviation of the signal. Correlation dimension measure chaotic signal complexity
for time-domain signal which is uniformly sampled is used in this work is specified
more clearly in [16].
2.3 Classification
Backpropagation neural network was used in this work. The number of hidden layers
used in this algorithm was set to 10. The maximum epoch was set to 1000. Tan-
Sigmoid was used as a transfer function. Levenberg Marquardt optimization is used
for this work. The validation method opted is k-fold where the value of k is set to 10.
Three parameters Accuracy, Sensitivity, and Specificity are calculated. Following
Eqs. (1–3) specifies the formulas. For the equations considered true positive cases
are where the interictal activity is correctly classified, true negative is one where
there is no actual activity and it is actually classified as such.
T r ue Positive
Sensitivity: = (2)
T r ue Positive + FalseN egative
T r ueN egative
Specificity: = (3)
T r ueN egative + False Positive
3 Results
Table 2 holds the results obtained using the methodology proposed in this work. For
this work, four cases have been evaluated for all of which accuracy, sensitivity, and
specificity have been calculated. Table 2 contains the results obtained. Highest accu-
racy of 99% has been achieved for case 2 (B–C) when comparing controlled group and
interictal activity recorded from epileptic zone; while comparing controlled activity
from and interictal activity from opposite from epileptic zone highest accuracy of
98.98% has been achieved for case 4 (B–D). This work indents on comparing inter-
ictal activity and activity from the controlled patients. For this work Backpropagation
neural is used to classify between them. Nonlinear entropy-based features approxi-
mate entropy, sample entropy, and fuzzy approximate entropy were extracted from
five sub-bands of both types of signals and correlation dimensions from a complete
signal.
Figures 3 and 4 compares correlation dimension of two sets used in this study from
which it can be observed that interictal activity which is represented by Set C and
Set D has higher correlation dimension than activity from controlled patient repre-
sented by Set A and Set B. Table 3 compares different studies on a similar database.
1.5
0.5
0
100
147
Set A Set C
1.5
0.5
100
147
Set B Set D
Previous researchers have worked on compared (a) activity from controlled patients
with epileptic, (b) activity from controlled patients with interictal, (c) interictal with
epileptic activity taking different cases. However, not much work is available to
compare activity from controlled patients with interictal activity. This study has
considered four such cases. From Table 3 it can be seen that different researchers
have worked on different cases on similar problems and the proposed methodology
outperforms [17–19] to the good margin where a different combination of sets for the
same problem is considered. Though, work done in [20, 21] have marginally outper-
formed the proposed method by 0.2% and 1% for case 1 and case 3 considered in
Automated Identification of Interictal Activity … 7
our work, respectively. However, not all cases have been considered in any of the
recent and previous studies.
4 Conclusion
To identify interictal discharges from routine EEG admits of artefacts through visual
inspection is a crucial but time-consuming task. To support the diagnosis of epilepsy
it is an important process. The automation of this process will ensure will supplement
the neurologist. This work is done to conquer the same problem. For validation k-
fold validation method is used in this work. We have considered four cases from the
publically available dataset and have achieved good performance measures.
Acknowledgements The authors express appreciation to R. G. Andrzejaket et al. for their public
accessible database [10].
References
1. Fisher RS, van Emde BW, Blume W et al (2005) Epileptic seizures and epilepsy: definitions
proposed by the International League against Epilepsy (ILAE) and the International Bureau
for Epilepsy (IBE). Epilepsia 46:470–472
2. Kobayashi T, Misaki K, Nakagawa H, Madokoro S, Ihara H, Tsuda K et al (1999) Non-linear
analysis of the sleep EEG. Psychiatry Clin Neurosci 53(2):159–161. https://doi.org/10.1046/j.
1440-1819.1999.00540
3. Lin R, Lee RG, Tseng CL, Zhou HK, Chao CF, Jiang JA (2006) A new approach for identifying
sleep apnea syndrome using wavelet transform and neural networks. Biomed Eng Appl Basis
Commun 18(3):138–144
4. Lin W-C, Chiu H-W, Hsu C-Y (2005) Discovering EEG signals response to musical signal
stimuli by time-frequency analysis and independent component analysis. In: Proceedings of
the 27th annual IEEE engineering in medicine and biology conference. Shanghai, China, pp
2765–2768
5. Babloyantz A, Nicolis C, Salazar JM (1985) Evidence of chaotic dynamics of brain activity
during the sleep cycle. Phys Lett 111A:152–157
6. Acharya UR et al (2012) Automated diagnosis of epileptic EEG using entropies. Biomed Signal
Process Control 7(4):401–408
7. Acharya UR, Fujita H, Sudarshan VK, Bhat S, Koh JEW (2015) Application of entropies for
automated diagnosis of epilepsy using EEG signals: a review. Knowledge-Based Syst 88:85–96
8. Tawfik NS, Youssef SM, Kholief M (2016) A hybrid automated detection of epileptic seizures
in EEG records. Comput Electr Eng 1(53):177–190
9. Acharya UR et al (2019) Characterization of focal EEG signals: a review. Future Gener Comput
Syst 91:290–299
10. Andrzejak RG, Lehnertz K, Mormann F, Rieke C, David P, Elger CE (2001) Indications
of nonlinear deterministic and finite-dimensional structures in time series of brain electrical
activity: dependence on recording region and brain state. Phys Rev E 64(6):061907
11. Kumar Y, Anand MLDRS (2014) Epileptic seizures detection in EEG using DWT-based ApEn
and artificial neural network, pp 1323–1334
8 A. Kaur et al.
12. Chen LL, Zhang J, Zou JZ, Zhao CJ, Wang GS (2014) A framework on wavelet-based nonlinear
features and extreme learning machine for epileptic seizure detection. Biomed Signal Process
Control 10(1):1–10
13. Xiang J, Li C, Li H, Cao R, Wang B, Han X, Chen J (2015) The detection of epileptic seizure
signals based on fuzzy entropy. J Neurosci Methods 243:18–25
14. Song Y, Liò P (2010) A new approach for epileptic seizure detection: sample entropy based
feature extraction and extreme learning machine. J Biomed Sci Eng 3(6):556–567
15. Bai D, Li X, Qiu T (2007) The sample entropy and its application in EEG based epilepsy
detection. J Biomed Eng 24(1):200–205
16. Theiler J (1987) Efficient algorithm for estimating the correlation dimension from a set of
discrete points. Phys Rev A 36(9):4456 American Physical Society
17. Sharma M, Pachori RB, Acharya UR (2017) A new approach to characterize epileptic seizures
using analytic time-frequency flexible wavelet transform and fractal dimension. Pattern Recogn
Lett 15(94):172–179
18. Hekim M (2016) The classification of EEG signals using discretization-based entropy and the
adaptive neuro-fuzzy inference system. Turkish J Electr Eng Comput Sci 24(1):285–297
19. Wang Y, Li Z, Feng L, Bai H, Wang C (2018) Hardware design of multiclass SVM classification
for epilepsy and epileptic seizure detection. IET Circ Dev Syst 12:108–115
20. Sriraam SRN (2017) Classification of epileptic seizures using wavelet packet log energy and
norm entropies with recurrent Elman neural network classifier. Cogn Neurodyn 11(1):51–66
21. Jaiswal AK, Banka H (2017) Local pattern transformation based feature extraction techniques
for classification of epileptic EEG signals. Biomed Signal Process Control 34:81–92. https://
doi.org/10.1016/j.bspc.2017.01.005
Random documents with unrelated
content Scribd suggests to you:
The Project Gutenberg eBook of La
Montserrat
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Title: La Montserrat
Novela de costums del nostre temps
Language: Catalan
Dolors Monserdà
1893
Comptava apenas dotze anys, quan los seus pares, pajesos dels
voltants de Lleyda, lo portaren á Barcelona, pera posarlo d’aprenent
en una d’eixas botigas molt en boga, per los anys de 1840,
anomenadas de mantegayre, en las que’ls amos, regularment fills de
Cerdanya, entremitg de mantas, flassadas, tapabocas, géneros de
punt y enfilalls de moradas y vermellas barretinas, hi venian
saborosa mantega, que directament rebian del seu país. Los
aprenents, sempre catalans, pero de montanya, entravan pera
menjar y dormir á la casa, en la que si eran honrats y aixerits,
regularment hi passavan los quatre anys del aprenentatje; sortintne
tant espabilats, com estalviadors y poch vanitosos, en lo relatiu á no
donarse vergonya de fer tot lo que creyan podia conduhirlos á la
realisació de sos ideals, que en aquella época de gustos moderats, se
reduhia senzillament á lo que se’n diu fer diners.
Dos anys feya que s’havia tret del convent á la més jove de las noyas
anomenada Adela, en lo moment en que aném á començar la nostra
narració. En Felip, lo fill més gran, ja de vint y cinch anys, havia fet lo
batxillerat en lo col·legi dels pares Jesuitas de Manresa, ahont havia
estat fins als disset, començant allavoras per gust propi la carrera
d’enginyer industrial, quins primers dos anys havia guanyat ab notas
de Notable y Sobresaliente; no aixís en los demés, ja que á l’época
que’l trobém feya tres anys que’s matriculava pe’l quart, del que
encara no havia eixit, gracias á las distraccions que’l voltavan y als
continuats viatjes efectuats al extranger, ahont havia passat molt
llargas temporadas.
—¡Ja ho pots ben dir! —saltá donya Francisqueta que desde feya
molts anys trobava sempre ben dit y ben pensat tot lo que s’ocorria al
seu espós.— ¡Jo’m marejo solament de véurho!
—Si, si, ’l teu pare tot vos ho compraria; pero quan veu los comptes ja
ho pensa de una altre manera!… y, ben fet que fá, —prosseguí Dª
Francisqueta adonantse de que sens pensar acabava de críticar al seu
marit— ¡Perqué si ho deixava á las vostras mans!… ¡Ay Deu meu! ¡Ja
caldrá que vos caseu ab un potentat, per qué sino!…
—¡Ben dit! ¡ben dit, noya! La qüestió son diners, diners y diners! Ab
lo temps tot passa… lo temps tot se’n ho emporta… y res, alló que
diuhen los castellans: Los duelos con pan son menos.
—Bé papá, pero á mi’m sembla, que en aixó també hi poden haber las
seuas excepcions —intervení l’Adela, abaixantse pera arreglar los
plechs de un matiné de batista, á fi d’amagar la rojor que se li habia
pujat á las galtas.— A voltas, hi ha joves, que si no tenen un gran
capital… tenen porvenir… que se’ls hi veu que farán carrera…
—¡Pe’l dot! ¡pe’l dot! ¿Y donchs, ximpleta? ¿Que’t pensas que’ls joves
no saben de comptar al temps que sóm? Me sembla noya, que tens
unas dérias que no son gayre del cas… En la vida real no hi ha més
que lo positiu. ¿Ho entens? tot lo demés son ximplesas. Mira, jo’m
vaig casar ab la teua mare, perque vaig coneixer que anava á fer una
cosa ben encertada. Jo’m vaig dir: Es una bona noya y te setze mil
duros lo dia que’s casará. Ella va pensar: es un bon jove y te una
galan botiga que’n val altres tants… Y dit y fet; nos varem casar. ¡Y ja
me’n rich de matrimonis que hajan viscut mellor que nosaltres!
—Tens rahó, noy. ¡Si ja es prop d’un quart de tres! —exclamá donya
Francisqueta, aixecantse darrera seu y cridant:— ¡Pepa! ¡Ramona!
¿Qué feu per aquesta cuyna? ¿No he dit que haviam de dinar á las
duas en punt? ¿No sabéu que havém de marxar en lo tren de las
cinch?
Donya Francisqueta se quedá com qui veu visions, en tant que don
Joaquím, que per ésser al altre extrém del cuarto, no havia entés las
paraulas del seu fill, pero que per la cara y actitut de la seua dona
comprengué que passava quelcom d’extraordinari, preguntá:
—Bé… Bé… ¿Es á dir que en lloch d’anar á Sant Hilari, com havias
dit, vens á Larrua? En fi, tu mateix: crech que ja tens edat de
sapiguer lo que’t convé… Y si no vas dret ¡pitjor per tu! Perque jo de
disbarats, no te’n deixaré fer ¿m’entens? Lo que no ha de ser, no será
¡Si qu’estariam frescos!… Si m’haguessis cregut á mi ¡qué’n fora de
diferenta la cosa!… ¡Quan penso que jo m’he tingut que fer tot sol… y
tu, un xicot que ab la teva posició, ó, mellor dit, ab la meva, —feu en
Bach repensantse— podias fer un magnífich casament… trobar un
primer dot!…
—Y bé, papá ¡no li passa pas la edat! ¡que’s diverteixi! ¡Si tots fan lo
mateix! —interposá l’Adela, desitjosa d’acabar aquell incident, que
habia vingut á rompre l’animació de la familia; y tot allargant una
cadira á don Joaquim, li passá carinyosament la má per sa grisenca
barba, dihentli á cau d’orella:
—¡Oh, l’Arenas es prou atent per haver anticipat lo sopar, pe’l gust de
saludar uns moments avans á las senyoretas Bach! —contestá la
interpelada: en tant que la Felissia deya mirant fixament al jove:
—De segur que l’Arenas, no te d’esperar á saludarlas en lo Balneari…
Lo natural es haverlas ja visitat eixa tarde…
—Mamá, per Deu ¡no digui aixó! Ellas podrán tenir com tothom las
sevas cosas ¡pero son molt bonas personas!
—Lo dir que fan olor de mantegayre, no vol significar que no sian
bonas personas, Encarnacioneta —saltá maliciosament la senyora
Gori.— Com vosté ha dit molt be, ellas podrán tenir las seuas
especialitats, pero ¡á bonas!… ¿No’s recorda, Paulina —afegí la nova
apologista de la familia de don Joaquím Bach, dirigintse á la senyora
Casabó— de lo molt que várem riure l’any passat per Sant Joaquím?
Perque, la veritat, donya Francisqueta es senzillota com ella sola ¡y
un colp llansada á explicar interioritats!…
La Clarita, la que feya deu dias tenia marejadas á la major part de las
senyoras de la colonia forastera, acabava d’entrar en lo saló, seguida
de son corpulent marit.
Foren las preguntas que feren á l’hora una munió de veus á donya
Antonieta, qui somriguent ab ayre de satisfacció y parlant ab més
calma quan més viva veya la curiositat del petit auditori, contestá:
—Es veritat.
—¡Te rahó!
—¡Qué poch coneix al pare Bach, si diu aixó! Jo li asseguro que quan
ha deixat pujar al noy es que aquesta qüestió ja no li inspira cap
temensa! —saltá ab seguretat la viuda Calcell.— ¡Los homes! ¡Los
homes! ¡Ells ray! ¡Aviat se’ls hi passan aquestas cosas!
—Qualsevol dia que tinga gust de sentir aquesta pessa, no te més que
passar un ratet per casa y faré que la noya la toqui pera que vosté
puga apreciar lo difícil y hermosa que es… ¡Si no comprench com no
s’equivoca un milió de vegadas ab aquest burgit, capás de fer perdre
al mateix Rubinstein!… Lo que sento, es que’l piano que tením aquí
es molt poca cosa… Un piano de torre ¿sap? Pero, vosté ja se’n fará
cárrech, y de tots modos, si á Barcelona nos fá l’obsequi de
visitarnos, li podrá sentir en lo piano de casa, …un magnífich Erard,
que’ns costá cinch cents duros. ¡Allí sí que aquesta fantasía hi fa un
preciosíssim efecte! Allí sí que hi lluheixen las dificultats vensudas,
l’agilitat de dits, y encara que me está mal dirho, la brillantor y
colorit de la execució…
—No entench com donya Narcisa, no coneix que la seua noya logrará
fernos apendre aquesta pessa de memoria! Fa cinch anys que vením
á Larrua, y ¡cinch anys que li sento tocar la mateixa fantasía! La