Groundwater analyses have revealed high concentrations of the toxic element arsenic around Stockholm and Mälardalen, a problem that often is linked to high levels of arsenic in the bedrock and which could be escalated by the many construction projects in the same region. However, it is unknown what part of the bedrock is causing the contamination. The aim of this thesis is to identify the chemical elements that associate with arsenic and study how the rock types differ in their content of elements and compounds. The highest median concentration of arsenic is found in quartz-feltspar-rich sedimentary rock, while intrusive rock types reveal the lowest levels. Using cluster analysis, arsenic is placed in a group including nine other elements, to which the strongest correlations are found with antimony, bismuth and silver. A moderate correlation with sulphur is also observed. The associations between groupings of elements are analysed using measures of dependence, which reveal relatively strong associations. Dimension reduction and ordination techniques provide further insight to the typical appearances of elements and reveal two groups of similar rock types.

No description available.

The aim of this thesis is to develop a method to predict consultation durations in a digital primary care setting and thereby create a tool for designing a more efficient scheduling system in primary care. The ultimate purpose of the work is to contribute to a reduction in waiting times in primary care. Even though no actual scheduling system was implemented, four machine learning models were implemented and compared to see if any of them had better performance. The input data used in this study was a combination of patient and doctor features. The patient features consisted of information extracted from digital symptom forms filled out by a patient before a video consultation with a doctor. These features were combined with doctor's speed, defined as the doctor's average consultation duration for his/her previous meetings. The output was defined as the length of the video consultation including administrative work made by the doctor before and after the meeting. One of the objectives of this thesis was to investigate whether the relationship between input and output was linear or non-linear. Also the problem was formulated both as a regression and a classification problem. The two problem formulations were compared in terms of achieved accuracy. The models chosen for this study was linear regression, linear discriminant analysis and the multi-layer perceptron implemented for both regression and classification. After performing a statistical t-test and a two-way ANOVA test it was concluded that no significant difference could be detected when comparing the models' performances. However, since linear regression is the least computationally heavy it was suggested for future usage until it is proved that any other model achieves better performance. Limitations such as too few models being tested and flaws in the data set were identified and further research is encouraged. Studies implementing an actual scheduling system using the methodology presented in the thesis is recommended as a topic for future research. / Syftet med denna uppsats är att utvärdera olika verktyg för att prediktera längden på ett läkarbesök och därmed göra det möjligt att skapa en mer effektiv schemaläggning i primärvården och på så sätt minska väntetiden för patienterna. Även om inget faktiskt schemaläggningssystem har föreslagits i denna uppsats så har fyra maskininlärningsmodeller implementerats och jämförts. Syftet med detta var bland annat att se om det var möjligt att dra slutsatsen att någon av modellerna gav bättre resultat än de andra. Den indata som använts i denna studie har bestått dels av symptomdata insamlad från symptomformulär ifylld av patienten före ett videomöte med en digital vårdgivare. Denna data har kombinerats med läkarens genomsnittliga mötestid i hens tidigare genomförda möten. Utdatan har definierats som längden av ett videomöte samt den tid som läkaren har behövt för administrativt arbete före och efter själva mötet. Ett av målen med denna studie var att undersöka som sambandet mellan indata och utdata är linjärt eller icke-linjärt. Ett annat mål var att formulera problemet både som ett regressionsproblem och som ett klassifikationsproblem. Syftet med detta var att kunna jämföra och se vilken av problemformuleringarna som gav bäst resultat. De modeller som har implementerats i denna studie är linjär regression, linjär diskriminationsanalys (linear discriminant analysis) och neurala nätverk implementerade för både regression och klassifikation. Efter att ha genomfört ett statistiskt t-test och en två-vägs ANOVA-analys kunde slutsatsen dras att ingen av de fyra studerade modellerna presterade signifikant bättre än någon av de andra. Eftersom linjär regression är enklare och kräver mindre datorkapacitet än de andra modellerna så dras slutsatsen att linjär regression kan rekommenderas för framtida användning tills det har bevisats att någon annan modell ger bättre resultat. De begränsningar som har identifierats hos studien är bland annat att det bara var fyra modeller som implementerats samt att datan som använts har vissa brister. Framtida studier som inkluderar fler modeller och bättre data har därför föreslagits. Dessutom uppmuntras framtida studier där ett faktiskt schemaläggningssystem implementeras som använder den metodik som föreslås i denna studie.

Improving national ranking is an increasingly important issue for university administrators. While research has been conducted on performance measures in higher education, research designs have lacked a predictive quality. Studies on the U.S. News college rankings have provided insight into the methodology; however, none of them have provided a model to predict what change in variable values would likely cause an institution to improve its standing in the rankings. The purpose of this study was to develop a predictive model for benchmarking academic programs (pBAP) for engineering colleges. The 2005 U.S. News ranking data for graduate engineering programs were used to create a four-tier predictive model (pBAP). The pBAP model correctly classified 81.9% of the cases in their respective tier. To test the predictive accuracy of the pBAP model, the 2005 U.S .News data were entered into the pBAP variate developed using the 2004 U.S. News data. The model predicted that 88.9% of the institutions would remain in the same ranking tier in the 2005 U.S. News rankings (compared with 87.7% in the actual data), and 11.1% of the institutions would demonstrate tier movement (compared with an actual 12.3% movement in the actual data). The likelihood of improving an institution's standing in the rankings was greater when increasing the values of 3 of the 11 variables in the U.S. News model: peer assessment score, recruiter assessment score, and research expenditures.

Laser Induced Breakdown Spectroscopy (LIBS) is showing great potential as an atomic analytical technique. With its ability to rapidly analyze all forms of matter, with little-to-no sample preparation, LIBS has many advantages over conventional atomic emission spectroscopy techniques. With the maturation of the technologies that make LIBS possible, there has been a growing movement to implement LIBS in portable analyzers for field applications. In particular, LIBS has long been considered the front-runner in the drive for stand-off detection of trace deposits of explosives. Thus there is a need for a better understanding of the relevant processes that are responsible for the LIBS signature and their relationships to the different system parameters that are helping to improve LIBS as a sensing technology. This study explores the use of LIBS as a method to detect random trace amounts of specific organic materials deposited on organic or non-metallic surfaces. This requirement forces the limitation of single-shot signal analysis. This study is both experimental and theoretical, with a sizeable component addressing data analysis using principal components analysis to reduce the dimensionality of the data, and quadratic discriminant analysis to classify the data. In addition, the alternative approach of ‘target factor analysis’ was employed to improve detection of organic residues on organic substrates. Finally, a new method of characterizing the laser-induced plasma of organics, which should lead to improved data collection and analysis, is introduced. The comparison between modeled and experimental measurements of plasma temperatures and electronic density is discussed in order to improve the present models of low-temperature laser induced plasmas.

Contamination of peanuts by secondary metabolites of certain fungi, namely aflatoxins present a great health hazard when exposed either at low levels for prolonged times (carcinogenic) or at high levels at once (poisonous). It is important to develop an accurate and rapid measurement technique to trace the aflatoxin and/or source fungi presence in peanuts. Thus, current research focused on development of vibrational spectroscopy based methods for detection and separation of contaminated peanut samples. Aflatoxin incidence, as a chemical contaminant in peanut paste samples, was investigated, in terms of spectral characteristics using FTIR-ATR. The effects of spectral pre-processing steps such as mean-centering, smoothing the 1st derivative and normalizing were studied. Logarithmic method was the best normalization technique describing the exponentially distributed spectral data. Spectral windows giving the best correlation with respect to increasing aflatoxin amount led to selection of fat associated spectral bands. Using the multivariate analysis tools, structural contributions of aflatoxins in peanut matrix were detected. The best region was decided as 3028-2752, 1800-1707, 1584-1424, and 1408-1127 cm-1 giving correlation coefficient for calibration (R2C), root mean square error for calibration (RMSEC) and root mean square error for prediction (RMSEP) of 98.6%, 7.66ppb and 19.5ppb, respectively. Applying the constructed partial least squares model, 95% of the samples were correctly classified while the percentage of false negative and false positive identifications were 16% and 0%, respectively. Aspergillus species of section Flavi and the black fungi, A. niger are the most common colonists of peanuts in nature and the majority of the aflatoxin producing strains are from section Flavi. Seed colonization by selected Aspergillus spp. was investigated by following the chemical alterations as a function of fungal growth by means of spectral readouts. FTIR-ATR was utilized to correlate spectral characteristics to mold density, and to separate Aspergillus at section, species and strain levels, threshold mold density values were established. Even far before the organoleptic quality changes became visually observable (~10,000 mold counts), FTIR distinguished the species of same section. Besides, the analogous secondary metabolites produced increased the similarity within the spectra even their spectral contributions were mostly masked by bulk peanut medium; and led to grouping of species producing the same mycotoxins together. Aflatoxigenic and non-aflatoxigenic strains of A. flavus and A. parasiticus were further studied for measurement capability of FTIR-ATR system in discriminating the toxic streams from just moldy and clean samples. Owing to increased similarity within the collected spectral data due to aflatoxin presence, clean samples (having aflatoxin level lower than 20 ppb, n=44), only moldy samples (having aflatoxin level lower than 300 ppb, n=28) and toxic samples (having aflatoxin level between 300-1200 ppb, n=23) were separated into appropriate classes (with a 100% classification accuracy). Photoacoustic spectroscopy (PAS) is a non-invasive technique and offers many advantages over more traditional ATR system, specifically, for in-field measurements. Even though the sample throughput time is longer compared to ATR measurements, intact seeds can be directly loaded into sample compartment for analysis. Compared to ATR, PAS is more sensitive to high moisture in samples, which in our case was not a problem since peanuts have water content less than 10%. The spectral ranges between: 3600-2750, 1800-1480, 1200-900 cm-1 were assigned as the key bands and full separation between Aspergillus spp. infected and healthy peanuts was obtained. However, PAS was not sensitive as ATR either in species level classification of Aspergillus invasion or toxic-moldy level separation. When run for separation of aflatoxigenic versus non-aflatoxigenic batches of samples, 7 out of 54 contaminated samples were misclassified but all healthy peanuts were correctly identified (15 healthy/ 69 total peanut pods). This study explored the possibility of using vibrational spectroscopy as a tool to understand chemical changes in peanuts and peanut products to Aspergillus invasion or aflatoxin contamination. The overall results of current study proved the potential of FTIR, equipped with either ATR or PAS, in identification, quantification and classification at varying levels of mold density and aflatoxin concentration. These results can be used to develop quality control laboratory methods or in field sorting devices. / Ph. D.

Institutions of higher education invest a significant amount of resources in recruiting, processing, and advising new students. When students leave the institution prior to graduation, the university loses considerable revenues. Therefore, it is important for colleges and universities to refine their student recruitment and retention strategies to avoid forgone revenues by predicting which students are likely to need particular types of support services (DeBerard et al, 2004). Current models of prediction utilize extensive surveys that are impractical to administer each term, and they do not adequately identify the broad range of student persistence categories needed in order to gain a greater understanding of persistence behavior (Davidson, 2005; Porter, 2000; Tinto, 1975). This study created a linear discriminant function to predict a broad range of persistence levels of first-time freshmen students at California State University, Bakersfield (CSUB), by identifying pre-enrollment and early enrollment student variables that existed within the database of the University. This information may be used to develop support service strategies to better assist incoming students predicted to have a greater probability of not persisting.

Wastewater treatment plants (WWTPs) face increasingly stringent effluent quality constraints as a result of rising environmental concerns. Efficient operation of the secondary clarification process is essential to be able to meet these strict regulations. Treatment plants can benefit greatly from making better use of available resources through improved automation and implementing more process systems engineering techniques to enhance plant performance. As such, the primary objective of this research is to utilize data-driven modeling techniques to obtain a representative model of a simplified secondary clarification unit in a WWTP. First, a deterministic subspace-based identification approach is used to estimate a linear state-space model of the secondary clarification process that can accurately predict process dynamics, with the ultimate objective of motivating the use of the subspace model in a model predictive control (MPC) framework for closed-loop control of the clarification process. To this end, a low-order subspace model which relates a set of typical measured outputs from a secondary clarifier to a set of typical inputs is identified and subsequently validated on simulated data obtained via Hydromantis's WWTP simulation software, GPS-X. Results illustrate that the subspace model is able to approximate the nonlinear process behaviour well and can effectively predict the dynamic output trajectory for various candidate input profiles, thus establishing its candidacy for use in MPC. Subsequently, a framework for forecasting the occurrence of sludge bulking--and consequently clarification failure--based on an engineered interaction variable that aims to capture the relationship between key input variables is proposed. Partial least squares discriminant analysis (PLS-DA) is used to discriminate between process conditions associated with clarification failure versus effective clarification. Preliminary results show that PLS-DA models augmented with the interaction variable demonstrate improved predictions and higher classification accuracy. / Thesis / Master of Applied Science (MASc)

No description available.

No description available.