For the effective qualitative management of drinking water, it is necessary to estimate the level of water pollution. In this research, to calculate the quality index of drinking water from the chemical parameters of Total Hardness, Alkalinity, Electrical Conductivity, Total Dissolved Solids, Calcium, Sodium, Magnesium, Potassium, Chlorine, Carbonate, Bicarbonate, and Sulfate in the hydrometric station of Bagh Kelayeh, Qazvin province used in the statistical period of 23 years (1998-2020). According to the calculated numerical values ​​and existing standards, water quality classified into two classes, good and excellent. To predict the quality class of drinking water based on chemical parameters, different combinations of parameters were considered in the form of several scenarios. In this regard, correlation and relief Algorithms were used to select different scenarios. Hoeffding Tree was used as a basic model for classifying water quality based on different combinations of parameters. Also, the performance of the combined Dagging approach in improving the results was evaluated. The results showed that the combined Dagging improves the water quality classification results. Scenario 6 Dagging with Hoeffding Tree base Algorithm, including HCO3, Ca, SO3, TDS, EC and TH parameters, with Kappa = 1, was introduced as the best method which is able to classify test samples correctly.

In this paper, we develop a non-visual automatic wrapper to extract data records from search engine results pages which contain important information for computer users. Our wrapper consists of a series of data filter to detect and remove irrelevant data from the web page. In the filtering stages, we incorporate two main Algorithms which are able to check the similarity of data records and to detect and extract the correct data region based on their component sizes. To evaluate the performance of our Algorithm, we carry out experimental and deletion tests. Experimental tests show that our wrapper outperforms the existing state of the art wrappers such as ViNT and DEPTA. Deletion studies by replacing our novel techniques with state of the art conventional techniques show that our wrapper design is efficient and could robustly extract data records from search engine results pages. With the speed advantages, our wrapper could be beneficial in processing large amount of web sites data, which could be helpful in meta search engine development.

Audio steganography is very important, equally steganography in other media (image, video, etc. ). In this paper is presented steganography of audio based on embedded zero-Tree waveleten transform Algorithm. Improving the resistance against white noise and additive noise with the lowest SNR is one of the important topics in steganography. The proposed Algorithm is more robust against normal white noise than uniform noise and has a bit error rate of less than 1 bit against SNRs higher than 10db. According to the obtained BER, if the proposed Algorithm is attacked, the hidden signal is lost completel. Also, the proposed method is resistant against additive noise. The proposed Algorithm has the least changes in the sound smoothness criterion in frequency domain with Capstrom distance scale and audio files in the form of music with soft tone (loudness), and the increase of secret message does not have much effect on creating disturbances in the frequency domain. The proposed Algorithm of the frequency spectrum does not change the audio signal much, and it also follows the property of the hearing threshold level, and high-pitched music with male speech has the best results, so it is favorable to the spectrum structure of Bark. Also, the proposed Algorithm has favorable results in the time domain. The lowest SNR is related to high-pitched music with female speech, which has an SNR of about 13db. According to the obtained results, we will have the worst case of embedding a secret message by choosing the audio signal with female speech. Because there is a certain smoothness in the fmale speech signal. Therefore, this uniform state will be lost to some extent by embedding a secret message in this type of audio signal, and the CZD criterion will increase according to the component-by-component comparison of the two main signals and the signal containing the secret message.

In this paper, we present a new method for improving the efficiency of indoor localization Algorithms, in terms of running time and error rate, using the KD-Tree data structure. One of the main challenges of indoor localization Algorithms in large environments is the high processing overhead of these Algorithms due to the high volume of input data and lack of processing resources in users' mobile devices. In the proposed method in this paper, we first cluster the fingerprint database. Then, with the help of a newly proposed method, a modified KD-Tree is implemented according to the conditions of the clusters. This Tree is a decision-making structure to select one specific cluster where the user stands there. Finally, when a user entered, using a few simple comparisons in the KD-Tree, the desired cluster is found and only information about that cluster is passed to the localization Algorithm, to compare and predict the user’s location. The results of the implementation of this method on the fingerprint data set of the Faculty of Engineering at Arak University show that the proposed method reduces the running times and errors to less than half the values, compared to the time of not using the proposed method.

The purpose of recent study is recognition of the most important financial ratios by which can evaluate company's performance. Therefore the total of accepted companies in stock exchange in Tehran which was active in 1390-1393 are considered as statistical universe of the research through which 102 companies organize the mass of statistical sample based on systematic elimination sampling method. The from the view point of exploratory and functional purpose, the research method is descriptive and interconnection including post – eventual researches. Analysis of data is accomplished by factorial analysis, structural equations modeling and two Algorithms by using of CHAID, C&RT software, SPSS, SMARTTPLS, CLEMENTIN decision Tree. after explanatory, factorial analysis, the results of research show that the number of 24 ratios from the total of considered 28 financial ratios is effective in evaluation of company's performance which these ratios are classified in seven categories in terms of weight of each of them from total variance by using of main factor analysis PCA. in next stage, for studying the type of relations and the amount of variant interconnection, confirmatory factorial analysis is performed in structural equations modeling and the main model presented. Finally the result and drawing decision Tree indicate that decision Tree Algorithms are presented the best prediction with the highest accuracy and among the sum of ratios, activity ratio has the most effect on performance evaluation.

Both fuzzy set theory and probability theory could handle uncertainty, and researchers have always compared the two to try to figure out which is better. Meanwhile, fuzzy decision Tree Algorithms represent classification knowledge more naturally to the way of human thinking. However, most generalizations of fuzzy decision Tree Algorithms focus only on numerical fine-tuning and interpretability of the Algorithms. In order to make full use of the information of data sets and improve the performance of fuzzy decision Trees. In this paper, we apply prior probability knowledge to the construction of fuzzy decision Tree. Therefore, first of all, based on prior probability knowledge, for a certain feature variable, we present a concept of feature value's class contribution level which shows the different roles the same feature values play  in different classes. Then, on the basis of feature value's class contribution level, we put forward a new Algorithm called Feature Contribution Fuzzy Decision Tree (FCFDT). FCFDT Algorithm has good classification results especially on data set with outliers. And it also maintains the interpretability of the fuzzy decision Tree Algorithms. The proposed Algorithm is implemented and validated with seven state-of-the-art decision Tree Algorithms on 15 real-life data sets coming from the UCI Repository of machine learning. The experimental results obtained clearly indicate the superiority of the proposed scheme over the baseline methods.