Open Source Data Mining Tool Research Articles

COMPARATIVE ANALYSIS OF OPEN-SOURCE DATA MINING TOOLS

COMPARATIVE ANALYSIS OF OPEN-SOURCE DATA MINING TOOLS

Data Mining Regarding Cyberbullying in the Arabic Language on Instagram Using KNIME and Orange Tools

This paper deals with data mining on verbal bullying by Instagram users. It tracks people who repeatedly have abusive behavior and may cause harm to other persons or groups. In this work, a dataset holding verbal bullying in the Arabic language was extracted from Instagram comments, and the entries were classified as regular verbal bullying and suspicious verbal bullying. KINIME and Orange open source data mining tools were utilized to discover comments that involved verbal bullying on Instagram and to delete previous comments while users sent their comments automatically and immediately. Classification algorithms Rule-Based in KNIME and Select Rows in Orange were used.

Vehicular Traffic in Urban Areas: Health Burden and Influence of Sustainable Urban Planning and Mobility

Vehicular traffic is one of the major sources of air pollution in European cities. This work aims to understand which characteristics of the urban environment could influence mobility-related air pollution, quantify the health impacts of exposure to traffic-derived PM2.5 and NO2 concentrations, and assess the potential health benefits expected from traffic interventions. The health benefits modeled are intended to provide a set of comparable data to support decision-makers and encourage informed decision-making to design healthier cities. Targeting a large geographical coverage, 12 European cities from 9 countries were comparatively assessed in terms of mean daily traffic volume/area, the number of public transport stops/area, and the percentage of green and outdoor leisure areas, among other urban indicators. This was implemented using an open-source data mining tool, which was seen as a useful engine to identify potential strategies to improve air quality. The comparison of urban indicators in the selected cities evidenced two trends: (a) cities with the most heterogeneous distribution of public transport stops, as an indicator of poor accessibility, are also those with the lowest proportion of km dedicated to cycleways and footways, highlighting the need in these cities for more sustainable mobility management; and (b) the percentage of green and outdoor leisure areas may influence the share of journeys by bicycle, pointing out that promoting the perception of green routes is relevant to enhance the potential of active transport modes. Socioeconomic factors can be key determinants of the urban indicators and would need further consideration. For the health impact assessment (HIA), two baseline scenarios were evaluated and compared. One is based on mean annual traffic contributions to PM2.5 concentrations in each target city (ranging between 1.9 and 13 µg/m3), obtained from the literature, and the second is grounded on mean annual NO2 concentrations at all available traffic and urban background stations within each city (17.2–83.5 µg/m3), obtained from the European Environment Agency database. The intervention scenarios modeled were designed based on traffic mitigation strategies in the literature, and set to ranges of 6–50% in traffic-derived PM2.5 concentrations and of 4–12.5% in NO2 concentrations. These scenarios could result in only a 1.7% (0.6–4%) reduction in premature mortality due to exposure to traffic-derived PM2.5, and 1.0% (0.4–2%) due to exposure to NO2, as the mean for all the cities. This suggests that more ambitious pollution abatement strategies should be targeted.

A Comprehensive Analysis of Proprietary and Open Source Data Mining Tools

The Powerful software tools and techniques required for the development of data mining applications. With the rapid development of technologies and business interest in using electronics and latest technologies plays important role in improvement of data mining field. Data mining access the meaningful and efficient information available in worldwide which is helps in decision making. This paper described the (a) various tools and techniques used by data mining applications. (b) compared features and limitations both in Proprietary and open sources data mining tools. (c) technical analysis of proprietary and open source data mining tools. On the basis of well-designed User interface, short time analysis, statistical and mathematical analysis user can select the best tool as per their requirements. Analysis of these tools makes easy to select appropriate tool.

Experimental Evaluation of Open Source Data Mining Tools: R, Rapid Miner and KNIME

In the current scenario of Big Data, open source Data Mining tools are very popular in business data analytics. The paper presents a comprehensive study of three most popular open source data mining tools – R, RapidMiner and KNIME. The tools are compared by implementing them on two real datasets. Performance is evaluated by creating a decision tree of the datasets taken. Our objective is to find the best tool for classification. The study can help researchers, developers and users in selecting a tool for accuracy in their data analysis and prediction. Experiments depict that accuracy level of the tool changes with the quantity and quality of the dataset. The results show that RapidMiner is the best tool followed by KNIME and R.

Pilot Study - Composite Risk Score for Pre-Transplant Risk Assessment in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Objective An internal pre-transplant risk assessment tool or Risk Monitor was developed by the Baylor Blood & Marrow Transplant (BMT) Data Management and Analytics team to assist with optimization of patient selection. The tool assigns a total risk score for allogeneic transplants based on attributes such as disease status, disease risk index (DRI), patient age, Karnofsky performance status (KPS), hematopoietic cellular therapy co-morbidity index (HCT-CI), and CMV status, along with psychosocial factors such as caregiver support, coping skills or motivation, health literacy, medication compliance, socioeconomic viability, distance of residence from the transplant center, as well as others. A retrospective pilot study was conducted to predict which attributes have the highest correlation with the primary outcome variable, post-transplant patient mortality. Method The Risk Monitor was developed using Microsoft Access (Microsoft®, Redmond, WA). A score of 0 (no risk) or 1 (risk) was assigned for each variable and an aggregate score was calculated for each patient (Figure 1). The data was analyzed using Waikato Environment for Knowledge Analysis (Weka). Weka is an open-source data mining tool consisting of a collection of machine learning algorithms for data mining tasks. It contains tools for data preparation, classification, regression, clustering, association rules mining, and visualization. Results We evaluated the total risk score for 97 patients that had undergone an allogeneic stem cell transplant (SCT) at Baylor University Medical Center in 2016. The total risk score for the patients varied between 0 and 9. Weka provided us with a list of independent attributes with rankings based on the intensity of correlation with the dependent variable, early mortality. Evaluation of the data set suggests that one attribute (Total Score) has the highest correlation with mortality, while other individual attributes (KPS, CMV status, DRI, HCT-CI and psychosocial factors such as caregiver support, coping skills, and health literacy) demonstrated a moderate to high correlation (Figure 2). Conclusion The Risk Monitor can serve as a useful tool for screening adult patients prior to allogeneic transplant. In addition, the tool may help to identify patients who need additional screening and social support to minimize the possibility of early mortality after transplantation. The Access database has been transitioned to a web-based (SQL Server/PHP/JavaScript) application to overcome certain Access-related issues such as client software installation, file corruption and interface update issues. Furthermore, this will enable easy cross platform usage, enhance security, and allow more concurrent users.

English

Data mining is the process of extracting informative and useful rules or relations, that can be used to make predictions about the values of new instances, from existing data. A wide range of commercial and open source software programs are used for data mining. In this study, a comparison of several classification algorithms included in some open source softwares such as WEKA, Tanagra and Scikit-learn using SEER (Survillance Epidemiology and End Results) data set which consists of 60948 instances is performed. Key words: Data mining, classification analysis, open source data mining tools. &nbsp

Learning from a Class Imbalanced Public Health Dataset: a Cost-based Comparison of Classifier Performance

Public health care systems routinely collect health-related data from the population. This data can be analyzed using data mining techniques to find novel, interesting patterns, which could help formulate effective public health policies and interventions. The occurrence of chronic illness is rare in the population and the effect of this class imbalance, on the performance of various classifiers was studied. The objective of this work is to identify the best classifiers for class imbalanced health datasets through a cost-based comparison of classifier performance. The popular, open-source data mining tool WEKA, was used to build a variety of core classifiers as well as classifier ensembles, to evaluate the classifiers’ performance. The unequal misclassification costs were represented in a cost matrix, and cost-benefit analysis was also performed. In another experiment, various sampling methods such as under-sampling, over-sampling, and SMOTE was performed to balance the class distribution in the dataset, and the costs were compared. The Bayesian classifiers performed well with a high recall, low number of false negatives and were not affected by the class imbalance. Results confirm that total cost of Bayesian classifiers can be further reduced using cost-sensitive learning methods. Classifiers built using the random under-sampled dataset showed a dramatic drop in costs and high classification accuracy.

An Improved Apriori Algorithm for Association Analysis

In recent years, association analysis has been gradually used in school teaching. In this paper, we proposed an improved NewApriori algorithm based on the Apriori algorithm. The New Apriori algorithm has combined the transaction data compression with the pruning of candidate item sets by using open source data mining tools weka. Three different transaction data of students scores are separately used on the experiment of NewApriori algorithm, the experiment result shows that the NewApriori algorithm is better than the original Apriori algorithm in time efficiency.

Evaluation and comparison of open source software suites for data mining and knowledge discovery

The growing interest in the extraction of useful knowledge from data with the aim of being beneficial for the data owner is giving rise to multiple data mining tools. Research community is specially aware of the importance of open source data mining software to ensure and ease the dissemination of novel data mining algorithms. The availability of these tools at no cost, and also the chance of better understanding of the approaches by examining their source code, provides the research community with an opportunity to tune and improve the algorithms. Documentation, updating, variety of algorithms, extensibility, and interoperability among others can be major issues to motivate users for opting for a specific open source data mining tool. The aim of this paper is to evaluate 19 open source data mining tools and to provide the research community with an extensive study based on a wide set of features that any tool should satisfy. The evaluation is carried out by following two methodologies. The first one is based on scores provided by experts to produce a subjective judgment of each tool. The second procedure performs an objective analysis about which features are satisfied by each tool. The ultimate aim of this work is to provide the research community with an extensive study on different features included in any data mining tool, either from a subjective and an objective point of view. Results reveal that RapidMiner, Konstanz Information Miner, and Waikato Environment for Knowledge Analysis are the tools that include higher percentage of these features. WIREs Data Mining Knowl Discov 2017, 7:e1204. doi: 10.1002/widm.1204This article is categorized under: Application Areas > Data Mining Software Tools Technologies > Computer Architectures for Data Mining

News Classification: A Data Mining Approach

Objectives: Text classification is one of the important applications of data mining. Text classification classifies text documents on the basis of words, phrases, combination of words etc. into predefined class labels. Method/Analysis: Present study classifies news data into four predefined classes namely Business, Entertainment, sports and Technology. For text classification WEKA an open source data mining tool is used. Different classification algorithms are applied on News data set. A comparative study of these algorithms is done based on Accuracy, Time, Errors and ROC to predict the best algorithm for news data set classification. Findings: Present study analyzed result on the basis of accuracy, time, error and ROC curve. Present work concludes that NaiveBayes Multinomial algorithm is best for news classification.

Predicting College Students Dropout using EDM Techniques

is study examines the factors affecting students' academic performance that contribute to the prediction of their failure and dropout using educational data mining techniques. This paper suggests the use of various classification techniques to identify the weak students who are likely to perform poorly in their academics. WEKA, an open source data mining tool was used to evaluate the attributes predicting student failure. The data set is comprised of 67 attributes of 150 students who have enrolled in B. Tech Degree Course registered for the academic year 2014-18 in a reputed college in Kerala affiliated to M.G University, Kerala, India. Various classification techniques like induction rules and decision tree have been applied to the data. The results of each of these approaches have been compared to select the one that achieves high accuracy.

A Study of Open Source Data Mining Tools and its Applications

Data Mining is a technology that is used for the process of analyzing and summarizing useful information from different perspectives of data. The importance of choosing data mining software tools for the developing applications using mining algorithms has led to the analysis of the commercially available open source data mining tools. The study discusses about the following likeness of the data mining tools-KNIME, WEKA, ORANGE, R Tool and Rapid Miner. The Historical Development and state-of-art; (i) The Applications supported, (ii) Data mining algorithms are supported by each tool, (iii) The pre-requisites and the procedures to install a tool, (iv) The input file format supported by each tool. To extract useful information from these data effectively and efficiently, data mining tools are used. The availability of many open source data mining tools, there is an increasing challenge in deciding upon the apace-updated tools for a given application. This study has provided a brief study about the open source knowledge discovery tools with their installation process, algorithms support and input file formats support.

EM&AA: An Algorithm for Predicting the Course Selection by Student in e-Learning Using Data Mining Techniques

Recommendation systems have been widely used in internet activities whose aim is to present the important and useful information to the user with little effort. Course Recommendation System is system which recommends to students the best combination of courses in engineering education system e.g. if student is interested in course like system programming then he would like to learn the course entitled compiler construction. The algorithm with combination of two data mining algorithm i.e. combination of Expectation Maximization Clustering and Apriori Association Rule Algorithm have been developed. The result of this developed algorithm is compared with Apriori Association Rule Algorithm which is an existing algorithm in open source data mining tool Weka.

Drug Discovery Applications for KNIME: An Open Source Data Mining Platform

Technological advances in high-throughput screening methods, combinatorial chemistry and the design of virtual libraries have evolved in the pursuit of challenging drug targets. Over the last two decades a vast amount of data has been generated within these fields and as a consequence data mining methods have been developed to extract key pieces of information from these large data pools. Much of this data is now available in the public domain. This has been helpful in the arena of drug discovery for both academic groups and for small to medium sized enterprises which previously would not have had access to such data resources. Commercial data mining software is sometimes prohibitively expensive and the alternate open source data mining software is gaining momentum in both academia and in industrial applications as the costs of research and development continue to rise. KNIME, the Konstanz Information Miner, has emerged as a leader in open source data mining tools. KNIME provides an integrated solution for the data mining requirements across the drug discovery pipeline through a visual assembly of data workflows drawing from an extensive repository of tools. This review will examine KNIME as an open source data mining tool and its applications in drug discovery.

Drug Discovery Applications for KNIME: An Open Source Data Mining Platform

Technological advances in high-throughput screening methods, combinatorial chemistry and the design of virtual libraries have evolved in the pursuit of challenging drug targets. Over the last two decades a vast amount of data has been generated within these fields and as a consequence data mining methods have been developed to extract key pieces of information from these large data pools. Much of this data is now available in the public domain. This has been helpful in the arena of drug discovery for both academic groups and for small to medium sized enterprises which previously would not have had access to such data resources. Commercial data mining software is sometimes prohibitively expensive and the alternate open source data mining software is gaining momentum in both academia and in industrial applications as the costs of research and development continue to rise. KNIME, the Konstanz Information Miner, has emerged as a leader in open source data mining tools. KNIME provides an integrated solution for the data mining requirements across the drug discovery pipeline through a visual assembly of data workflows drawing from an extensive repository of tools. This review will examine KNIME as an open source data mining tool and its applications in drug discovery.

Selecting the Best Supervised Learning Algorithm for Recommending the Course in E-Learning System

a day E-learning is becoming popular as it helps to fulfil the necessities of remote students and helps the teaching-learning process in Education system. Course Recommender System in E-Learning is a system which recommend the course to the student based on the choice of various student collected from huge amount of data of courses offered through Moodle package of the college. Here in this paper we compare the seven classification algorithm to choose the best classification algorithm for Course Recommendation system. Theses seven classification algorithms are ADTree, Simple Cart, J48, ZeroR, Naive Bays, Decision Table & Random Forest Classification Algorithm. We compare these seven algorithms using open source data mining tool Weka & present the result. We found that ADTree classification algorithm works better for this Course Recommender System than other five classification algorithms. KeywordsSimple Cart, J48, ZeroR, Naive Bays, Decision Table, Random Forest Classification Algorithm, Weka

Data Preparation Strategy in E-Learning System using Association Rule Algorithm

Data preparation is the important step in Course Recommendation System which aims at predicting the course selected by student. In this paper we present the data preparation strategy for Course Recommendation System. Here we have used the real data of courses offered through Moodle package of the college & apply the data preparation strategy to it to find out the best combination of courses. We are using the open source data mining tool Weka to check the result.

Prediction of Course Selection by Student using Combination of Data Mining Algorithms in E-Learning

recommender system aims at predicting the best combination of courses selected by students. Here in this paper we present how the combination of clustering algorithm- Simple K-means Algorithm & association rule algorithm- Apriori Association Rule is useful in Course Recommender system. If we use only the Apriori association rule algorithm then we need to preprocess the data obtained from Moodle database. But if we use this combination of clustering & association rule then there is no need to preprocess the data. So we present this new approach & also present the result. To test the result we have used the open source data mining tool Weka.

Ranking of Influencing Factors in Predicting Students’ Academic Performance

The aim of this study was to rank influencing factors that contribute to the prediction of students’ academic performance. It is useful in identifying weak students who are likely to perform poorly in their studies. In this study, we used WEKA open source data mining tool to analyze attributes for predicting a higher learning institution’s bachelor of computer science students’ academic performance. The data set comprised of 2427 number of student records and 396 attributes of students registered between year 2000 and 2006. Preprocessing includes attribute importance analysis. We applied the data set to different classifiers (Bayes, trees or function) and obtained the accuracy of predicting the students’ performance into either first-second-upper class or second-lower-third class. A cross-validation with 10 folds was used to evaluate the prediction accuracy. Our results showed the ranking of courses that has significant impact on predicting the students’ overall academic results. In addition, we perform experiments comparing the performance of different classifiers and the result showed that Naive Bayes, AODE and RBFNetwork classifiers scored the highest percentage of prediction accuracy of 95.29%.