Aquesta tesi versa sobre la classificació automàtica de gèneres musicals, basada en l'anàlisi del contingut del senyal d'àudio, plantejant-ne els problemes i proposant solucions. Es proposa un estudi de la classificació de gèneres musicals des del punt de vista computacional però inspirat en teories dels camps de la musicologia i de la percepció. D'aquesta manera, els experiments presentats combinen diferents elements que influeixen en l'encert o fracàs de la classificació, com ara els descriptors d'àudio, les tècniques d'aprenentatge, etc. L'objectiu és avaluar i comparar els resultats obtinguts d'aquests experiments per tal d'explicar els límits d'encert dels algorismes actuals, i proposar noves estratègies per tal de superar-los. A més a més, partint del processat de la informació d'àudio, s'inclouen aspectes musicals i culturals referents al gènere que tradicionalment no han estat tinguts en compte en els estudis existents. En aquest context, es proposa l'estudi de diferents famílies de descriptors d'àudio referents al timbre, ritme, tonalitat o altres aspectes de la música. Alguns d'aquests descriptors són proposats pel propi autor mentre que d'altres ja són perfectament coneguts. D'altra banda, també es comparen les tècniques d'aprenentatge artificial que s'usen tradicionalment en aquest camp i s'analitza el seu comportament davant el nostre problema de classificació. També es presenta una discussió sobre la seva capacitat per representar els diferents models de classificació proposats en el camp de la percepció. Els resultats de la classificació es comparen amb un seguit de tests i enquestes realitzades sobre un conjunt d'individus. Com a resultat d'aquesta comparativa es proposa una arquitectura específica de classificadors que també està raonada i explicada en detall. Finalment, es fa un especial èmfasi en comparar resultats dels classificadors automàtics en diferents escenaris que pressuposen la barreja de bases de dades, la comparació entre bases de dades grans i petites, etc. A títol de conclusió, es mostra com l'arquitectura de classificació proposada, justificada pels resultats dels diferents anàlisis, pot trencar el límit actual en tasques de classificació automàtica de gèneres musicals. De manera condensada, es pot dir que aquesta tesi contribueix al camp de la classificació de gèneres musicals en els següents aspectes: a) Proporciona una revisió multidisciplinar delsgèneres musicals i la seva classificació; b)Presenta una avaluació qualitativa i quantitativa de les famílies de descriptors d'àudio davant el problema de la classificació de gèneres; c) Avalua els pros i contres de les diferents tècniques d'aprenentatge artificial davant el gènere; d) Proposa una arquitectura nova de classificador d'acord amb una visió interdisciplinar dels gèneres musicals; e) Analitza el comportament de l'arquitecturaproposada davant d'entorns molt diversos en el que es podria implementar el classificador. / Esta tesis estudia la clasificación automática degéneros musicales, basada en el análisis delcontenido de la señal de audio, planteando sus problemas y proponiendo soluciones. Sepropone un estudio de la clasificación de los géneros musicales desde el punto de vista computacional, pero inspirado en teorías de los campos de la musicología y la percepción. De este modo, los experimentos persentados combinan distintos elementos que influyen en el acierto o fracaso de la clasificación, como por ejemplo los descriptores de audio, las técnicas de aprondiza je, etc. El objetivo es comparar y evaluar los resultados obtenidos de estos experimentos para explicar los límites de las tasas de acierto de los algorismos actuales, y proponer nuevas estrategias para superarlos. Además, partiendo del procesado de la información de Audio, se han incluido aspectos musicales y culturales al género que tradicionalmente no han sido tomados en cuenta en los estudios existentes. En este contexto, se propone el estudio de distintas famílias de descriptores de audio referentes al timbre, al ritmo, a la tonalidad o a otros aspectos de la música. Algunos de los descriptores son propuestos por el mismo autor, mientras que otros son perfectamente conocidos. Por otra parte, también se comparan las técnicas de aprendiza je artificial que se usan tradicionalmente, y analizamos su comportamiento en frente de nuestro problema de clasificación. Tambien planteamos una discusión sobre su capacidad para representar los diferentes modelos de clasificación propuestos en el campo de la percepción. Estos resultados de la clasificación se comparan con los resultados de unos tests y encuestas realizados sobre un conjunto de individuos. Como resultado de esta comparativa se propone una arquitectura específica de clasificadores que tambien está razonada y detallada en el cuerpo de la tesis. Finalmente, se hace un émfasis especial en comparar los resultados de los clasificadores automáticos en distintos escenarios que assumen la mezcla de bases de datos, algunas muy grandes y otras muy pequeñas, etc. Como conclusión, mostraremos como la arquitectura de clasificación propuesta permite romper el límite actual en el ámbito de la classificación automática de géneros musicales.De forma condensada, se puede decir que esta tesis contribuye en el campo de la clasificación de los géneros musicales el los siguientes aspectos: a) Proporciona una revisión multidisciplinar de los géneros musicales y su clasificación; b) Presenta una evaluación cualitativa y cuantitativa de las famílias de descriptores de audio para la clasificación de géneros musicales; c) Evalua los pros y contras de las distintas técnicas de aprendiza je artificial delante del género; d) Propone una arquitectura nueva del clasificador de acuerdo con una visión interdisciplinar de los géneros musicales; e) Analiza el comportamiento de la arquitectura propuesta delante de entornos muy diversos en los que se podria implementar el clasificador. / This dissertation presents, discusses, and sheds some light on the problems that appear when computers try to automatically classify musical genres from audio signals. In particular, a method is proposed for the automatic music genre classification by using a computational approach that is inspired in music cognition and musicology in addition to Music Information Retrieval techniques. In this context, we design a set of experiments by combining the different elements that may affect the accuracy in the classification (audio descriptors, machine learning algorithms, etc.). We evaluate, compare and analyze the obtained results in order to explain the existing glass-ceiling in genre classification, and propose new strategies to overcome it. Moreover, starting from the polyphonic audio content processing we include musical and cultural aspects of musical genre that have usually been neglected in the current state of the art approaches. This work studies different families of audio descriptors related to timbre, rhythm, tonality and other facets of music, which have not been frequently addressed in the literature. Some of these descriptors are proposed by the author and others come from previous existing studies. We also compare machine learning techniques commonly used for classification and analyze how they can deal with the genre classification problem. We also present a discussion on their ability to represent the different classification models proposed in cognitive science. Moreover, the classification results using the machine learning techniques are contrasted with the results of some listening experiments proposed. This comparison drive us to think of a specific architecture of classifiers that will be justified and described in detail. It is also one of the objectives of this dissertation to compare results under different data configurations, that is, using different datasets, mixing them and reproducing some real scenarios in which genre classifiers could be used (huge datasets). As a conclusion, we discuss how the classification architecture here proposed can break the existing glass-ceiling effect in automatic genre classification. To sum up, this dissertation contributes to the field of automatic genre classification: a) It provides a multidisciplinary review of musical genres and its classification; b) It provides a qualitative and quantitative evaluation of families of audio descriptors used for automatic classification; c) It evaluates different machine learning techniques and their pros and cons in the context of genre classification; d) It proposes a new architecture of classifiers after analyzing music genre classification from different disciplines; e) It analyzes the behavior of this proposed architecture in different environments consisting of huge or mixed datasets.

Made available in DSpace on 2015-05-14T13:21:38Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 6875549 bytes, checksum: 3697064e0b5c3d3ac90181f954575bc7 (MD5) Previous issue date: 2013-06-21 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Tea has an economic and cultural importance, not only for producers and consumers, but also for a scientific interest. The organoleptic quality of the Camellia sinensis infusion depends on the nature and amount of several secondary metabolites (such as polyphenols, caffeine, amino acids, etc.), which can be directly related to the geographical origin of the tea plants. These components are the basis of the economic value of teas and its beneficial effects on human health. Therefore, there is a growing consumer s interest in high quality teas with a distinct geographical identity. In last decades, the analytical methods employing modern instrumental techniques have become more sensitive, reliable and fast. However, these techniques have advantages and limitations for the application in the analyses of the tea quality and their geographic origins. Thus, a combination of different techniques could be more useful than relying on a single method. Following these principles, we propose three new strategies for simultaneous classification of teas according to both the type (green and black) and geographic origin (Argentina, Brazil and Sri Lanka). The proposed methodologies employ the use of (1) digital images, (2) NIR spectroscopy, and (3) chemical composition (moisture, ash, caffeine, total polyphenols, fluoride and fifteen metals (Na, Mg, Al, P, K, Ca, Cr, Mn, Fe, Co, Ni, Cu, Zn, Cd and Pb) in both tea leaves and infusions). A correct classification of all tea samples (100% of correct classification) was always obtained using the Linear Discriminant Analysis associated with the variable selection technique taken by the Successive Projections Algorithm. Soft Independent Modeling of Class Analogy (SIMCA) and Partial Least Squares Discriminant Analysis (PLS-DA) were also used. The proposed strategies might be useful for the development of legislation for the quality control of teas in Brazil, which is still lacking / O chá tem uma importância econômica e cultural, não só para produtores e consumidores, mas também por um interesse científico. A qualidade organoléptica da infusão da Camellia sinensis depende da natureza e da quantidade de vários metabólitos secundários (tais como polifenóis, cafeína, aminoácidos, etc.), os quais podem ser relacionados diretamente com a origem geográfica das plantas. Estes componentes são a base do valor econômico do chá e de seus efeitos benéficos sobre a saúde humana. Por isso, há um crescente interesse dos consumidores por chás de alta qualidade com uma clara identidade geográfica. Durante as últimas décadas, as metodologias analíticas que empregam técnicas instrumentais modernas tornaram-se mais sensíveis, confiáveis e rápidas. Entretanto, tais técnicas têm vantagens e limitações para a aplicação da análise da qualidade do chá e de suas origens geográficas. Assim, uma combinação de diferentes técnicas analíticas pode ser mais útil do que depender de um único método. Seguindo estes preceitos, nós propusemos três novas estratégias para a classificação simultânea de chás de acordo com o tipo (verde e preto) e a origem geográfica (Argentina, Brasil e Sri Lanka). As metodologias propostas empregam o uso de (1) imagens digitais, (2) espectroscopia NIR e (3) composição química (umidade, cinza total, cafeína, polifenóis totais, fluoreto e quinze metais (Na, Mg, Al, P, K, Ca, Cr, Mn, Fe, Co, Ni, Cu, Zn, Cd e Pb) nas folhas e infusões dos chás). Uma classificação correta de todas as amostras de chás (100% de acerto) foi sempre obtida utilizando Análise Discriminante Linear associada à técnica de seleção de variáveis feita pelo Algoritmo das Projeções Sucessivas (SPA-LDA). Modelagem Independente e Flexível por Analogia de Classe (SIMCA) e Análise Discriminante por Mínimos Quadrados Parciais (PLS-DA) também foram utilizadas. Tais estratégias podem ser úteis para a elaboração de normas para o controle de qualidade de chás no Brasil, que ainda é inexistente

The topic of this Master’s thesis is combination of laser spectroscopic methods. LIBS and Raman spectroscopy were chosen for the combination. This combination is applied to plastic identification and separation as a mean to automate sorting of plastic waste. Data handling was done via different methods of computer learning algorithms scripted in R language. Plastic sorting accuracy over 90 % was reached thanks to the combination of chosen methods. This work also addresses some issues implied by combination of two different methods.

A regional survey of environmentally harmful substances in fish in autumn 2013 revealed elevated levels of polychlorinated biphenyls (PCBs) in perch from Lake Oxundasjön, north of Upplands Väsby in Stockholm County. Follow-up studies have shown that the quantity of PCBs contained in the lake is unique of its kind in Sweden, and that the area of influence also includes downstream Rosersbergsviken, a bay of Lake Mälaren. The elevated concentrations in fish exceed today's market limits and environmental quality standards for PCBs, as of this, responsible authorities discourage from consumption of fish from Lake Oxundasjön and Rosersbergsviken. The aim has been to use statistical analyses and mass-balance modelling to study the current state of the lake system and how it may evolve in the future under different circumstances. There is a statistically significant correlation between PCB levels in sediment and perch from 21 different sites in the Stockholm-Mälaren region, including Lake Oxundasjön and Rosersbergsviken, this was demonstrated with a linear regression model. With the multivariate analysis method principal component analysis (PCA), it was illustrated how the contaminant levels in fish from Lake Oxundasjön and Rosersbergsviken differed on contaminant levels in fish from other sites. The difference mainly concerned the size and composition of PCBs. Mass-balance modelling of quantities and flows of PCBs in Lake Oxundasjön and Rosersbergsviken was made in the simulation program STELLA®. The modelling indicated that the system currently serves as a secondary distribution source of PCBs to the environment. The recovery of PCB levels is slow in the system, it will take more than 25 years for concentrations in fish to reach today’s market limits and environmental quality standards for PCBs. The model was used to evaluate three different treatment methods for Lake Oxundasjön: dredging, capping and activated carbon treatment. Simulations of these treatments led to a substantial improvement of the PCB situation in Lake Oxundasjön. Moreover, they also had a positive impact on the recovery process in the downstream Rosersbergsviken. Future climate changes, with warmer temperatures and higher run off, led to a slightly faster recovery progress of PCBs in the system.

Human hair profiling is an integral part of a forensic investigation but it is one of the most technically difficult subjects in forensic science. This thesis describes the research and development of a novel approach for the rapid identification of unknown human and other related keratin fibres found at a crime scene. The work presented here is developed systematically and considers sample collection, sample preparation, analysis and interpretation of spectral data for the profiling of hair fibres encountered in criminal cases. Spectral comparison of fibres was facilitated with the use of chemometrics methods such as PCA, SIMCA and Fuzzy Clustering, and the less common approach of multi-criteria decision making methodology (MCDM). The aim of the thesis was to investigate the potential of some vibrational spectroscopy techniques for matching and discrimination of single keratin hair fibres in the context of forensic evidence. The first objective (chapter 3) of the thesis was to evaluate the use of Raman and FT-IR micro-spectroscopy techniques for the forensic sampling of hair fibres and to propose the preferred technique for future forensic hair comparisons. The selection of the preferred technique was based on criteria such as spectral quality, ease of use, rapid analysis and universal application to different hair samples. FT-IR micro-spectroscopy was found to be the most appropriate technique for hair analysis because it enabled the rapid collection of spectra from a wide variety of hair fibres. Raman micro-spectroscopy, on the other hand, was hindered with fluorescence problems and did not allow the collection of spectra from pigmented fibres. This objective has therefore shown that FT-IR micro-spectroscopy is the preferable spectroscopic technique for forensic analysis of hair fibres, whilst Raman spectroscopy is the least preferred. The second objective (chapter 3) was to investigate, through a series of experiments, the effect of chemical treatment on the micro-environment of human hair fibres. The effect of bleaching agents on the hair fibres was studied with some detail at different treatment times and the results indicate a significant change in the chemical environment of the secondary structure of the hair fibre along with changes in the C-C backbone structure. One of the most important outcomes of this research was the behaviour of the fÑ-helix during chemical treatment. The hydrogen bonding in the fÑ-helix provides for the stable structure of the fibre and therefore any disruption to the fÑ-helix will inevitably damage the molecular structure of the fibre. The results highlighted the behaviour of the fÑ-helix, which undergoes a significant decrease in content during oxidation, and is partly converted to a random-coil structure, whilst the fÒ-sheet component of the secondary structure remains unaffected. The reported investigations show that the combination of FT-IR and Raman micro-spectroscopy can provide an insight and understanding into the complex chemical properties and reactions within a treated hair fibre. Importantly, this work demonstrates that with the aid of chemometrics, it is possible to investigate simultaneously FT-IR and Raman micro-spectroscopic information from oxidised hair fibres collected from one subject and treated at different times. The discrimination and matching of hair fibres on the basis of treatment has potential forensic applications. The third objective (chapter 4) attempted to expand the forensic application of FT-IR micro-spectroscopy to other keratin fibres. Animal fibres are commonly encountered in crime scenes and it thus becomes important to establish the origin of those fibres. The aim of this work was to establish the forensic applications of FT-IR micro-spectroscopy to animal fibres and to investigate any fundamental molecular differences between these fibres. The results established a discrimination between fibres consisting predominantly of fÑ-helix and those containing mainly a fÒ-sheet structure. More importantly, it was demonstrated through curve-fitting and chemometrics, that each keratin fibre contains a characteristic secondary structure arrangement. The work presented here is the first detailed FT-IR micro-spectroscopic study, utilising chemometrics as well as MCDM methods, for a wide range of keratin fibres, which are commonly, found as forensic evidence. Furthermore, it was demonstrated with the aid of the rank ordering MCDM methods PROMETHEE and GAIA, that it is possible to rank and discriminate keratin fibres according to their molecular characteristics obtained from direct measurements together with information sourced from the literature. The final objective (chapter 5) of the thesis was to propose an alternative method for the discrimination and matching of single scalp human hair fibres through the use of FT-IR micro-spectroscopy and chemometrics. The work successfully demonstrated, through a number of case scenarios, the application of the technique for the identification of variables such as gender and race for an unknown single hair fibre. In addition, it was also illustrated that known hair fibres (from the suspect or victim) can be readily matched to the unknown hair fibres found at the crime scene. This is the first time that a substantial, systematic FT-IR study of forensic hair identification has been presented. The research has shown that it is possible to model and correlate individual¡¦s characteristics with hair properties at molecular level with the use of chemometrics methods. A number of different, important forensic variables of immediate use to police in a crime scene investigation such as gender, race, treatment, black and white hair fibres were investigated. Blind samples were successfully applied both to validate available experimental data and extend the current database of experimental determinations. Protocols were posed for the application of this methodology in the future. The proposed FT-IR methodology presented in this thesis has provided an alternative approach to the characterisation of single scalp human hair fibres. The technique enables the rapid collection of spectra, followed by the objective analytical capabilities of chemometrics to successfully discriminate animal fibres, human hair fibres from different sources, treated from untreated hair fibres, as well as black and white hair fibres, on the basis of their molecular structure. The results can be readily produced and explained in the courts of law. Although the proposed relatively fast FT-IR technique is not aimed at displacing the two slower existing methods of hair analysis, namely comparative optical microscopy and DNA analysis, it has given a new dimension to the characterisation of hair fibres at a molecular level, providing a powerful tool for forensic investigations.

Fire debris analysis currently relies on visual pattern recognition of the total ion chromatograms, extracted ion profiles, and target compound chromatograms to identify the presence of an ignitable liquid. This procedure is described in the ASTM International E1618-10 standard method. For large data sets, this methodology can be time consuming and is a subjective method, the accuracy of which is dependent upon the skill and experience of the analyst. This research aimed to develop an automated classification method for large data sets and investigated the use of the total ion spectrum (TIS). The TIS is calculated by taking an average mass spectrum across the entire chromatographic range and has been shown to contain sufficient information content for the identification of ignitable liquids. The TIS of ignitable liquids and substrates were compiled into model data sets. Substrates are defined as common building materials and household furnishings that are typically found at the scene of a fire and are, therefore, present in fire debris samples. Fire debris samples were also used which were obtained from laboratory-scale and large-scale burns. An automated classification method was developed using computational software that was written in-house. Within this method, a multi-step classification scheme was used to detect ignitable liquid residues in fire debris samples and assign these to the classes defined in ASTM E1618-10. Classifications were made using linear discriminant analysis, quadratic discriminant analysis (QDA), and soft independent modeling of class analogy (SIMCA). The model data sets iv were tested by cross-validation and used to classify fire debris samples. Correct classification rates were calculated for each data set. Classifier performance metrics were also calculated for the first step of the classification scheme which included false positive rates, true positive rates, and the precision of the method. The first step, which determines a sample to be positive or negative for ignitable liquid residue, is arguably the most important in the forensic application. Overall, the highest correct classification rates were achieved using QDA for the first step of the scheme and SIMCA for the remaining steps. In the first step of the classification scheme, correct classification rates of 95.3% and 89.2% were obtained using QDA to classify the crossvalidation test set and fire debris samples, respectively. For this step, the cross-validation test set resulted in a true positive rate of 96.2%, a false positive rate of 9.3%, and a precision of 98.2%. The fire debris data set had a true positive rate of 82.9%, a false positive rate of 1.3%, and a precision of 99.0%. Correct classifications rates of 100% were achieved for both data sets in the majority of the remaining steps which used SIMCA for classification. The lowest correct classification rate, 69.2%, was obtained for the fire debris samples in one of the final steps in the classification scheme. In this research, the first statistically valid error rates for fire debris analysis have been developed through cross-validation of large data sets. The fire debris analyst can use the automated method as a tool for detecting and classifying ignitable liquid residues in fire debris samples. The error rates reduce the subjectivity associated with the current methods and provide a level of confidence in sample classification that does not currently exist in forensic fire debris analysis.

No description available.

<p>In this thesis some multivariate spectroscopic methods for the analysis of solutions are proposed. Spectroscopy and multivariate data analysis form a powerful combination for obtaining both quantitative and qualitative information and it is shown how spectroscopic techniques in combination with chemometric data evaluation can be used to obtain rapid, simple and efficient analytical methods. These spectroscopic methods consisting of spectroscopic analysis, a high level of automation and chemometric data evaluation can lead to analytical methods with a high analytical capacity, and for these methods, the term high-capacity analysis (HCA) is suggested. It is further shown how chemometric evaluation of the multivariate data in chromatographic analyses decreases the need for baseline separation. </p><p>The thesis is based on six papers and the chemometric tools used are experimental design, principal component analysis (PCA), soft independent modelling of class analogy (SIMCA), partial least squares regression (PLS) and parallel factor analysis (PARAFAC). The analytical techniques utilised are scanning ultraviolet-visible (UV-Vis) spectroscopy, diode array detection (DAD) used in non-column chromatographic diode array UV spectroscopy, high-performance liquid chromatography with diode array detection (HPLC-DAD) and fluorescence spectroscopy. The methods proposed are exemplified in the analysis of pharmaceutical solutions and serum proteins.</p><p>In Paper I a method is proposed for the determination of the content and identity of the active compound in pharmaceutical solutions by means of UV-Vis spectroscopy, orthogonal signal correction and multivariate calibration with PLS and SIMCA classification. Paper II proposes a new method for the rapid determination of pharmaceutical solutions by the use of non-column chromatographic diode array UV spectroscopy, i.e. a conventional HPLC-DAD system without any chromatographic column connected. In Paper III an investigation is made of the ability of a control sample, of known content and identity to diagnose and correct errors in multivariate predictions something that together with use of multivariate residuals can make it possible to use the same calibration model over time. In Paper IV a method is proposed for simultaneous determination of serum proteins with fluorescence spectroscopy and multivariate calibration. Paper V proposes a method for the determination of chromatographic peak purity by means of PCA of HPLC-DAD data. In Paper VI PARAFAC is applied for the decomposition of DAD data of some partially separated peaks into the pure chromatographic, spectral and concentration profiles. </p>

In this thesis some multivariate spectroscopic methods for the analysis of solutions are proposed. Spectroscopy and multivariate data analysis form a powerful combination for obtaining both quantitative and qualitative information and it is shown how spectroscopic techniques in combination with chemometric data evaluation can be used to obtain rapid, simple and efficient analytical methods. These spectroscopic methods consisting of spectroscopic analysis, a high level of automation and chemometric data evaluation can lead to analytical methods with a high analytical capacity, and for these methods, the term high-capacity analysis (HCA) is suggested. It is further shown how chemometric evaluation of the multivariate data in chromatographic analyses decreases the need for baseline separation. The thesis is based on six papers and the chemometric tools used are experimental design, principal component analysis (PCA), soft independent modelling of class analogy (SIMCA), partial least squares regression (PLS) and parallel factor analysis (PARAFAC). The analytical techniques utilised are scanning ultraviolet-visible (UV-Vis) spectroscopy, diode array detection (DAD) used in non-column chromatographic diode array UV spectroscopy, high-performance liquid chromatography with diode array detection (HPLC-DAD) and fluorescence spectroscopy. The methods proposed are exemplified in the analysis of pharmaceutical solutions and serum proteins. In Paper I a method is proposed for the determination of the content and identity of the active compound in pharmaceutical solutions by means of UV-Vis spectroscopy, orthogonal signal correction and multivariate calibration with PLS and SIMCA classification. Paper II proposes a new method for the rapid determination of pharmaceutical solutions by the use of non-column chromatographic diode array UV spectroscopy, i.e. a conventional HPLC-DAD system without any chromatographic column connected. In Paper III an investigation is made of the ability of a control sample, of known content and identity to diagnose and correct errors in multivariate predictions something that together with use of multivariate residuals can make it possible to use the same calibration model over time. In Paper IV a method is proposed for simultaneous determination of serum proteins with fluorescence spectroscopy and multivariate calibration. Paper V proposes a method for the determination of chromatographic peak purity by means of PCA of HPLC-DAD data. In Paper VI PARAFAC is applied for the decomposition of DAD data of some partially separated peaks into the pure chromatographic, spectral and concentration profiles.