This paper presents a novel approach of using machine learning algorithms based on experts’ knowledge to classify web pages into three predefined classes according to the degree of content adjustment to the search engine optimization (SEO) recommendations. In this study, classifiers were built and trained to classify an unknown sample (web page) into one of the three predefined classes and to identify important factors that affect the degree of page adjustment. The data in the training set are manually labeled by domain experts. The experimental results show that machine learning can be used for predicting the degree of adjustment of web pages to the SEO recommendations—classifier accuracy ranges from 54.59% to 69.67%, which is higher than the baseline accuracy of classification of samples in the majority class (48.83%). Practical significance of the proposed approach is in providing the core for building software agents and expert systems to automatically detect web pages, or parts of web pages, that need improvement to comply with the SEO guidelines and, therefore, potentially gain higher rankings by search engines. Also, the results of this study contribute to the field of detecting optimal values of ranking factors that search engines use to rank web pages. Experiments in this paper suggest that important factors to be taken into consideration when preparing a web page are page title, meta description, H1 tag (heading), and body text—which is aligned with the findings of previous research. Another result of this research is a new data set of manually labeled web pages that can be used in further research.

Growing the Ross broiler parent according to the target growth curve ensures that males and females achieve optimum lifetime performance and well-being. Accurate control of growth will lead to uniformity and sexual maturity, which are of crucial importance for the production of hygienic, healthy, and fertile eggs of high quality. This study examined the growth of Ross 308 broiler breeder flocks from hatch to 35 wk of age to identify which growth model would describe the growth of these animals most accurately. Growth was measured and modeled using linear and nonlinear functions, and the experimental growth curves were compared with target curves from the Parent Stock Management Manual for Ross 308 (Aviagen). Broiler breeder flock R6 (in-season from February until October) and flock R7 (off-season from August until April) were kept in an environmentally controlled breeder house from hatch until 35 wk of age. Three nonlinear growth functions (logistic, Gompertz, and Richards) and 3 polynomial functions (linear, second-order, and third-order) were applied. Parameters of the models were estimated by the least squares procedure. The fit of growth curves to experimental data was assessed using R(2). A t-test was used to identify significant differences in the goodness of fit of the model to the different data sets (breeder manual, R6, and R7). The third-order polynomial gave the best fit to the Ross 308 parent broiler BW data, with R(2) ranging from 0.992 to 0.998. Among the nonlinear growth functions, the Richards model gave the best fit to the data, with R(2) ranging from 0.992 to 0.995. The advantage of second- and higher-order polynomial models is that they can be linearized and their parameters estimated by linear regression.

The botanical origin of starch is of importance in industrial applications and food processing because it may influence the properties of the final product. Current microscopic methods are time-consuming. Starch consists of an origin-dependent amylose/amylopectin ratio. Triiodide ions bind characteristically to the amylose and amylopectin depending on the botanical origin of the starch. The absorbance of the starch-triiodide complex was measured for wheat, potato, corn, rye, barley, rice, tapioca and unknown origin starch, and within the different cultivars. Each starch sample had specific parameters: starch-triiodide complex peak wavelength maximum (λmax/nm), maximum absorbance change at λmax (ΔA) and λmax shift towards the unknown origin starch sample values. The visible absorption spectra (500-800 nm) for each starch sample were used as a unique fingerprint, and then elaborated by cluster analysis. The cluster analysis managed to distinguish data of two clusters, a cereal type cluster and a potato/tapioca/rice starch cluster. The cereal subclusters extensively distinguished wheat/barley/rye starches from corn starches. Data for cultivars were mostly in good agreement within the same subclaster. The proposed method that combines cluster analysis and visible absorbance data for starch-triiodide complex was able to distinguish starch of different botanical origins and cultivars within the same species. This method is simpler and more convenient than standard time-consuming methods.

Macro‐ and microelements in the samples of virgin and cold pressed pumpkin seed oils produced in Croatia through two consecutive crop seasons were determined by inductively coupled plasma–optical emission spectroscopy (ICP‐OES). Croatian oils were also compared to oils from Slovenia and Austria in order to assess differences in the element content. Magnesium, potassium, calcium, sodium, selenium, and iron were the dominant elements in all pumpkin seed oils. Their amounts together with barium, strontium, manganese, copper were up to ninefold higher ( p ≤ 0.05) in virgin compared to cold pressed pumpkin seed oils. These differences occur due to the different processing conditions which include salt addition, heat treatment, and higher degree of equipment ware out during virgin pumpkin seed oil production. As the sodium level increases with the addition of salt, virgin pumpkin seed oil could be considered its hidden source and producers should pay attention to the amount added. Contents of cobalt, copper, selenium, and thallium significantly differed ( p ≤ 0.05) between the two crop seasons. Principal component analysis revealed clear differences between samples with different origin that can be explained by the specifics in the production processes of each country. In comparison with Austrian and Slovenian, Croatian pumpkin seed oils had significantly lower contents of sodium, potassium, calcium, magnesium, and tin while bismuth and selenium were higher.

Text classification is an important and common task in supervised machine learning. The Naive Bayes Classifier is a popular algorithm that can be used for this purpose. The goal of our research was prediction of song performer using Naive Bayes classification algorithm based solely on lyrics. A dataset that has been created consists of lyrics performed by Nirvana and Metallica, 207 songs in total. Model evaluation measures showed very good results: precision of 0.93, recall of 0.95 and F 1 -measure of 0.94, therefore lyrics classification using Naive Bayes can be considered as successful.

No abstract

In the research, in this paper, we investigate spatial and temporal variations in the composition of wastewater near Croatian highways in three climatic regions (continental, Mediterranean, highland) during three seasons (autumn, winter and spring). In our paper, the spatial division of the investigated areas that pertain to the three aforementioned climatic regions was obtained using the method of hierarchical clustering of monitored locations. One thousand five hundred thirty-three samples from 14 locations along Croatian highways were collected and analysed by methods of multivariate exploratory analysis. By methods of principal components, factor analysis and hierarchical clustering of variables, we grouped the variables into factors. Whereas 60 % of variation in the data was explained by three principal components, six principal components accounted for 88 % of data variation. The key section of our research was conducted by the decision tree method. For the purpose of analysis, we classified 1,533 samples into three classes representing climatic regions separately for each season and obtained the accuracy of 76-90 % on test samples. Finally, using decision trees, we identified the most important variables that differentiate climatic regions by the level of contamination of water along highways in different seasons.

In this paper we have analysed concetrations of heavy metals (lead, copper, nickel, zink, mercury, cadmium, and chromium) in wastewater along highways in Croatia. We have used standard statistical methods: analysis of variance, Kruskal-Wallis test and principal analysis. Analysis of variance and Kruskal-Wallis test were used to detect factors that influence the concentration of lead, copper, nickel, and zink in wastewater. We have investigated the influence of the highway sampling location, the side of a highway, and the influence of the season of the year. Principal components were used to identify groups of elements with similar characteristics in wastewater.