LScD (Leicester Scientific Dictionary)April 2020 by Neslihan Suzen, PhD student at the University of Leicester (ns433@leicester.ac.uk/suzenneslihan@hotmail.com)Supervised by Prof Alexander Gorban and Dr Evgeny Mirkes[Version 3] The third version of LScD (Leicester Scientific Dictionary) is created from the updated LSC (Leicester Scientific Corpus) - Version 2*. All pre-processing steps applied to build the new version of the dictionary are the same as in Version 2** and can be found in description of Version 2 below. We did not repeat the explanation. After pre-processing steps, the total number of unique words in the new version of the dictionary is 972,060. The files provided with this description are also same as described as for LScD Version 2 below.* Suzen, Neslihan (2019): LSC (Leicester Scientific Corpus). figshare. Dataset. https://doi.org/10.25392/leicester.data.9449639.v2** Suzen, Neslihan (2019): LScD (Leicester Scientific Dictionary). figshare. Dataset. https://doi.org/10.25392/leicester.data.9746900.v2[Version 2] Getting StartedThis document provides the pre-processing steps for creating an ordered list of words from the LSC (Leicester Scientific Corpus) [1] and the description of LScD (Leicester Scientific Dictionary). This dictionary is created to be used in future work on the quantification of the meaning of research texts. R code for producing the dictionary from LSC and instructions for usage of the code are available in [2]. The code can be also used for list of texts from other sources, amendments to the code may be required.LSC is a collection of abstracts of articles and proceeding papers published in 2014 and indexed by the Web of Science (WoS) database [3]. Each document contains title, list of authors, list of categories, list of research areas, and times cited. The corpus contains only documents in English. The corpus was collected in July 2018 and contains the number of citations from publication date to July 2018. The total number of documents in LSC is 1,673,824.LScD is an ordered list of words from texts of abstracts in LSC.The dictionary stores 974,238 unique words, is sorted by the number of documents containing the word in descending order. All words in the LScD are in stemmed form of words. The LScD contains the following information:1.Unique words in abstracts2.Number of documents containing each word3.Number of appearance of a word in the entire corpusProcessing the LSCStep 1.Downloading the LSC Online: Use of the LSC is subject to acceptance of request of the link by email. To access the LSC for research purposes, please email to ns433@le.ac.uk. The data are extracted from Web of Science [3]. You may not copy or distribute these data in whole or in part without the written consent of Clarivate Analytics.Step 2.Importing the Corpus to R: The full R code for processing the corpus can be found in the GitHub [2].All following steps can be applied for arbitrary list of texts from any source with changes of parameter. The structure of the corpus such as file format and names (also the position) of fields should be taken into account to apply our code. The organisation of CSV files of LSC is described in README file for LSC [1].Step 3.Extracting Abstracts and Saving Metadata: Metadata that include all fields in a document excluding abstracts and the field of abstracts are separated. Metadata are then saved as MetaData.R. Fields of metadata are: List_of_Authors, Title, Categories, Research_Areas, Total_Times_Cited and Times_cited_in_Core_Collection.Step 4.Text Pre-processing Steps on the Collection of Abstracts: In this section, we presented our approaches to pre-process abstracts of the LSC.1.Removing punctuations and special characters: This is the process of substitution of all non-alphanumeric characters by space. We did not substitute the character “-” in this step, because we need to keep words like “z-score”, “non-payment” and “pre-processing” in order not to lose the actual meaning of such words. A processing of uniting prefixes with words are performed in later steps of pre-processing.2.Lowercasing the text data: Lowercasing is performed to avoid considering same words like “Corpus”, “corpus” and “CORPUS” differently. Entire collection of texts are converted to lowercase.3.Uniting prefixes of words: Words containing prefixes joined with character “-” are united as a word. The list of prefixes united for this research are listed in the file “list_of_prefixes.csv”. The most of prefixes are extracted from [4]. We also added commonly used prefixes: ‘e’, ‘extra’, ‘per’, ‘self’ and ‘ultra’.4.Substitution of words: Some of words joined with “-” in the abstracts of the LSC require an additional process of substitution to avoid losing the meaning of the word before removing the character “-”. Some examples of such words are “z-test”, “well-known” and “chi-square”. These words have been substituted to “ztest”, “wellknown” and “chisquare”. Identification of such words is done by sampling of abstracts form LSC. The full list of such words and decision taken for substitution are presented in the file “list_of_substitution.csv”.5.Removing the character “-”: All remaining character “-” are replaced by space.6.Removing numbers: All digits which are not included in a word are replaced by space. All words that contain digits and letters are kept because alphanumeric characters such as chemical formula might be important for our analysis. Some examples are “co2”, “h2o” and “21st”.7.Stemming: Stemming is the process of converting inflected words into their word stem. This step results in uniting several forms of words with similar meaning into one form and also saving memory space and time [5]. All words in the LScD are stemmed to their word stem.8.Stop words removal: Stop words are words that are extreme common but provide little value in a language. Some common stop words in English are ‘I’, ‘the’, ‘a’ etc. We used ‘tm’ package in R to remove stop words [6]. There are 174 English stop words listed in the package.Step 5.Writing the LScD into CSV Format: There are 1,673,824 plain processed texts for further analysis. All unique words in the corpus are extracted and written in the file “LScD.csv”.The Organisation of the LScDThe total number of words in the file “LScD.csv” is 974,238. Each field is described below:Word: It contains unique words from the corpus. All words are in lowercase and their stem forms. The field is sorted by the number of documents that contain words in descending order.Number of Documents Containing the Word: In this content, binary calculation is used: if a word exists in an abstract then there is a count of 1. If the word exits more than once in a document, the count is still 1. Total number of document containing the word is counted as the sum of 1s in the entire corpus.Number of Appearance in Corpus: It contains how many times a word occurs in the corpus when the corpus is considered as one large document.Instructions for R CodeLScD_Creation.R is an R script for processing the LSC to create an ordered list of words from the corpus [2]. Outputs of the code are saved as RData file and in CSV format. Outputs of the code are:Metadata File: It includes all fields in a document excluding abstracts. Fields are List_of_Authors, Title, Categories, Research_Areas, Total_Times_Cited and Times_cited_in_Core_Collection.File of Abstracts: It contains all abstracts after pre-processing steps defined in the step 4.DTM: It is the Document Term Matrix constructed from the LSC[6]. Each entry of the matrix is the number of times the word occurs in the corresponding document.LScD: An ordered list of words from LSC as defined in the previous section.The code can be used by:1.Download the folder ‘LSC’, ‘list_of_prefixes.csv’ and ‘list_of_substitution.csv’2.Open LScD_Creation.R script3.Change parameters in the script: replace with the full path of the directory with source files and the full path of the directory to write output files4.Run the full code.References[1]N. Suzen. (2019). LSC (Leicester Scientific Corpus) [Dataset]. Available: https://doi.org/10.25392/leicester.data.9449639.v1[2]N. Suzen. (2019). LScD-LEICESTER SCIENTIFIC DICTIONARY CREATION. Available: https://github.com/neslihansuzen/LScD-LEICESTER-SCIENTIFIC-DICTIONARY-CREATION[3]Web of Science. (15 July). Available: https://apps.webofknowledge.com/[4]A. Thomas, "Common Prefixes, Suffixes and Roots," Center for Development and Learning, 2013.[5]C. Ramasubramanian and R. Ramya, "Effective pre-processing activities in text mining using improved porter’s stemming algorithm," International Journal of Advanced Research in Computer and Communication Engineering, vol. 2, no. 12, pp. 4536-4538, 2013.[6]I. Feinerer, "Introduction to the tm Package Text Mining in R," Accessible en ligne: https://cran.r-project.org/web/packages/tm/vignettes/tm.pdf, 2013.

The Supplementary Material of the paper "Supplementary Material: Predictive model using Cross Industry Standard Process for Data Mining" includes: 1) APPENDIX 1: SQL Statements for data extraction. Appendix 2: Interview for operating Staff. 2) The DataSet of the normalized data to define the predictive model.

This data collection is associated with the publications: Merdith, A. S., Landgrebe, T. C. W., Dutkiewicz, A., & Müller, R. D. (2013). Towards a predictive model for opal exploration using a spatio-temporal data mining approach. Australian Journal of Earth Sciences, 60(2), 217-229. doi: 10.1080/08120099.2012.754793

and

Landgrebe, T. C. W., Merdith, A., Dutkiewicz, A., & Müller, R. D. (2013). Relationships between palaeogeography and opal occurrence in Australia: A data-mining approach. Computers & Geosciences, 56(0), 76-82. doi: 10.1016/j.cageo.2013.02.002

Publication Abstract - Merdith et al. (2013)

Opal is Australia's national gemstone, however most significant opal discoveries were made in the early 1900's - more than 100 years ago - until recently. Currently there is no formal exploration model for opal, meaning there are no widely accepted concepts or methodologies available to suggest where new opal fields may be found. As a consequence opal mining in Australia is a cottage industry with the majority of opal exploration focused around old opal fields. The EarthByte Group has developed a new opal exploration methodology for the Great Artesian Basin. The work is based on the concept of applying “big data mining” approaches to data sets relevant for identifying regions that are prospective for opal. The group combined a multitude of geological and geophysical data sets that were jointly analysed to establish associations between particular features in the data with known opal mining sites. A “training set” of known opal localities (1036 opal mines) was assembled, using those localities, which were featured in published reports and on maps. The data used include rock types, soil type, regolith type, topography, radiometric data and a stack of digital palaeogeographic maps. The different data layers were analysed via spatio-temporal data mining combining the GPlates PaleoGIS software (www.gplates.org) with the Orange data mining software (orange.biolab.si) to produce the first opal prospectivity map for the Great Artesian Basin. One of the main results of the study is that the geological conditions favourable for opal were found to be related to a particular sequence of surface environments over geological time. These conditions involved alternating shallow seas and river systems followed by uplift and erosion. The approach reduces the entire area of the Great Artesian Basin to a mere 6% that is deemed to be prospective for opal exploration. The work is described in two companion papers in the Australian Journal of Earth Sciences and Computers and Geosciences.

Publication Abstract - Landgrebe et al. (2013)

Age-coded multi-layered geological datasets are becoming increasingly prevalent with the surge in open-access geodata, yet there are few methodologies for extracting geological information and knowledge from these data. We present a novel methodology, based on the open-source GPlates software in which age-coded digital palaeogeographic maps are used to “data-mine” spatio-temporal patterns related to the occurrence of Australian opal. Our aim is to test the concept that only a particular sequence of depositional/erosional environments may lead to conditions suitable for the formation of gem quality sedimentary opal. Time-varying geographic environment properties are extracted from a digital palaeogeographic dataset of the eastern Australian Great Artesian Basin (GAB) at 1036 opal localities. We obtain a total of 52 independent ordinal sequences sampling 19 time slices from the Early Cretaceous to the present-day. We find that 95% of the known opal deposits are tied to only 27 sequences all comprising fluvial and shallow marine depositional sequences followed by a prolonged phase of erosion. We then map the total area of the GAB that matches these 27 opal-specific sequences, resulting in an opal-prospective region of only about 10% of the total area of the basin. The key patterns underlying this association involve only a small number of key environmental transitions. We demonstrate that these key associations are generally absent at arbitrary locations in the basin. This new methodology allows for the simplification of a complex time-varying geological dataset into a single map view, enabling straightforward application for opal exploration and for future co-assessment with other datasets/geological criteria. This approach may help unravel the poorly understood opal formation process using an empirical spatio-temporal data-mining methodology and readily available datasets to aid hypothesis testing.

Authors and Institutions

Andrew Merdith - EarthByte Research Group, School of Geosciences, The University of Sydney, Australia. ORCID: 0000-0002-7564-8149

Thomas Landgrebe - EarthByte Research Group, School of Geosciences, The University of Sydney, Australia

Adriana Dutkiewicz - EarthByte Research Group, School of Geosciences, The University of Sydney, Australia

R. Dietmar Müller - EarthByte Research Group, School of Geosciences, The University of Sydney, Australia. ORCID: 0000-0002-3334-5764

Overview of Resources Contained

This collection contains geological data from Australia used for data mining in the publications Merdith et al. (2013) and Landgrebe et al. (2013). The resulting maps of opal prospectivity are also included.

List of Resources

Note: For details on the files included in this data collection, see “Description_of_Resources.txt”.

Note: For information on file formats and what programs to use to interact with various file formats, see “File_Formats_and_Recommended_Programs.txt”.

• Map of Barfield region, Australia (.jpg, 270 KB)
• Map overviewing the Great Artesian basins and main opal mining camps (.png, 82 KB)
• Maps showing opal prospectivity data mining results for different geological datasets (.tif, 23.1 MB)
• Map of opal prospectivity from palaeogeography data mining (.pdf, 2.6 MB)
• Raster of palaeogeography target regions for viewing in Google Earth (.jpg, 418 KB)
• Opal mine locations (.gpml, .txt, .kmz, .shp, total 15.6 MB)
• Map of opal prospectivity from all data mining results as a Google Earth overlay (.kmz, 12 KB)
• Map of probability of opal occurrence in prospective regions from all data mining results (.tif, 5.9 MB)
• Paleogeography of Australia (.gpml, .txt, .shp, total 114.2 MB)
• Radiometric data showing potassium concentration contrasts (.tif, .kmz, total 311.3 MB)
• Regolith data (.gpml, .txt, .kml, .shp, total 7.1 MB)
• Soil type data (.gpml, .txt, .kml, .shp, total 7.1 MB)

For more information on this data collection, and links to other datasets from the EarthByte Research Group please visit EarthByte

For more information about using GPlates, including tutorials and a user manual please visit GPlates or EarthByte

In this research, we have generated student retention alerts. The alerts are classified into two types: preventive and corrective. This classification varies according to the level of maturity of the data systematization process. Therefore, to systematize the data, data mining techniques have been applied. The experimental analytical method has been used, with a population of 13,715 students with 62 sociological, academic, family, personal, economic, psychological, and institutional variables, and factors such as academic follow-up and performance, financial situation, and personal information. In particular, information is collected on each of the problems or a combination of problems that could affect dropout rates. Following the methodology, the information has been generated through an abstract data model to reflect the profile of the dropout student. As advancement from previous research, this proposal will create preventive and corrective alternatives to avoid dropout higher education. Also, in contrast to previous work, we generated corrective warnings with the application of data mining techniques such as neural networks until reaching a precision of 97% and losses of 0.1052. In conclusion, this study pretends to analyze the behavior of students who drop out the university through the evaluation of predictive patterns. The overall objective is to predict the profile of student dropout, considering reasons such as admission to higher education and career changes. Consequently, using a data systematization process promotes the permanence of students in higher education. Once the profile of the dropout has been identified, student retention strategies have been approached, according to the time of its appearance and the point of view of the institution.

This file contains the life cycle inventories (LCIs) developed for an associated journal article. Potential users of the data are referred to the journal article for a full description of the modeling methodology. LCIs were developed for cumene and sodium hydroxide manufacturing using data mining with metadata-based data preprocessing. The inventory data were collected from US EPA's 2012 Chemical Data Reporting database, 2011 National Emissions Inventory, 2011 Toxics Release Inventory, 2011 Electronic Greenhouse Gas Reporting Tool, 2011 Discharge Monitoring Report, and the 2011 Biennial Report generated from the RCRAinfo hazardous waste tracking system. The U.S. average cumene gate-to-gate inventories are provided without (baseline) and with process allocation applied using metadata-based filtering. In 2011, there were 8 facilities reporting public production volumes of cumene in the U.S., totaling to 2,609,309,687 kilograms of cumene produced that year. The U.S. average sodium hydroxide gate-to-gate inventories are also provided without (baseline) and with process allocation applied using metadata-based filtering. In 2011, there were 24 facilities reporting public production volumes of sodium hydroxide in the U.S., totaling to 3,878,021,614 kilograms of sodium hydroxide produced that year. Process allocation was only conducted for the top 12 facilities producing sodium hydroxide, which represents 97% of the public production of sodium hydroxide. The data have not been compiled in the formal Federal Commons LCI Template to avoid users interpreting the template to mean the data have been fully reviewed according to LCA standards and can be directly applied to all types of assessments and decision needs without additional review by industry and potential stakeholders. This dataset is associated with the following publication: Meyer, D.E., S. Cashman, and A. Gaglione. Improving the reliability of chemical manufacturing life cycle inventory constructed using secondary data. JOURNAL OF INDUSTRIAL ECOLOGY. Berkeley Electronic Press, Berkeley, CA, USA, 25(1): 20-35, (2021).

The purpose of data mining analysis is always to find patterns of the data using certain kind of techiques such as classification or regression. It is not always feasible to apply classification algorithms directly to dataset. Before doing any work on the data, the data has to be pre-processed and this process normally involves feature selection and dimensionality reduction. We tried to use clustering as a way to reduce the dimension of the data and create new features. Based on our project, after using clustering prior to classification, the performance has not improved much. The reason why it has not improved could be the features we selected to perform clustering are not well suited for it. Because of the nature of the data, classification tasks are going to provide more information to work with in terms of improving knowledge and overall performance metrics. From the dimensionality reduction perspective: It is different from Principle Component Analysis which guarantees finding the best linear transformation that reduces the number of dimensions with a minimum loss of information. Using clusters as a technique of reducing the data dimension will lose a lot of information since clustering techniques are based a metric of 'distance'. At high dimensions euclidean distance loses pretty much all meaning. Therefore using clustering as a "Reducing" dimensionality by mapping data points to cluster numbers is not always good since you may lose almost all the information. From the creating new features perspective: Clustering analysis creates labels based on the patterns of the data, it brings uncertainties into the data. By using clustering prior to classification, the decision on the number of clusters will highly affect the performance of the clustering, then affect the performance of classification. If the part of features we use clustering techniques on is very suited for it, it might increase the overall performance on classification. For example, if the features we use k-means on are numerical and the dimension is small, the overall classification performance may be better. We did not lock in the clustering outputs using a random_state in the effort to see if they were stable. Our assumption was that if the results vary highly from run to run which they definitely did, maybe the data just does not cluster well with the methods selected at all. Basically, the ramification we saw was that our results are not much better than random when applying clustering to the data preprocessing. Finally, it is important to ensure a feedback loop is in place to continuously collect the same data in the same format from which the models were created. This feedback loop can be used to measure the model real world effectiveness and also to continue to revise the models from time to time as things change.

The LSC (Leicester Scientific Corpus)

April 2020 by Neslihan Suzen, PhD student at the University of Leicester (ns433@leicester.ac.uk) Supervised by Prof Alexander Gorban and Dr Evgeny MirkesThe data are extracted from the Web of Science [1]. You may not copy or distribute these data in whole or in part without the written consent of Clarivate Analytics.[Version 2] A further cleaning is applied in Data Processing for LSC Abstracts in Version 1*. Details of cleaning procedure are explained in Step 6.* Suzen, Neslihan (2019): LSC (Leicester Scientific Corpus). figshare. Dataset. https://doi.org/10.25392/leicester.data.9449639.v1.Getting StartedThis text provides the information on the LSC (Leicester Scientific Corpus) and pre-processing steps on abstracts, and describes the structure of files to organise the corpus. This corpus is created to be used in future work on the quantification of the meaning of research texts and make it available for use in Natural Language Processing projects.LSC is a collection of abstracts of articles and proceeding papers published in 2014, and indexed by the Web of Science (WoS) database [1]. The corpus contains only documents in English. Each document in the corpus contains the following parts:1. Authors: The list of authors of the paper2. Title: The title of the paper 3. Abstract: The abstract of the paper 4. Categories: One or more category from the list of categories [2]. Full list of categories is presented in file ‘List_of _Categories.txt’. 5. Research Areas: One or more research area from the list of research areas [3]. Full list of research areas is presented in file ‘List_of_Research_Areas.txt’. 6. Total Times cited: The number of times the paper was cited by other items from all databases within Web of Science platform [4] 7. Times cited in Core Collection: The total number of times the paper was cited by other papers within the WoS Core Collection [4]The corpus was collected in July 2018 online and contains the number of citations from publication date to July 2018. We describe a document as the collection of information (about a paper) listed above. The total number of documents in LSC is 1,673,350.Data ProcessingStep 1: Downloading of the Data Online

The dataset is collected manually by exporting documents as Tab-delimitated files online. All documents are available online.Step 2: Importing the Dataset to R

The LSC was collected as TXT files. All documents are extracted to R.Step 3: Cleaning the Data from Documents with Empty Abstract or without CategoryAs our research is based on the analysis of abstracts and categories, all documents with empty abstracts and documents without categories are removed.Step 4: Identification and Correction of Concatenate Words in AbstractsEspecially medicine-related publications use ‘structured abstracts’. Such type of abstracts are divided into sections with distinct headings such as introduction, aim, objective, method, result, conclusion etc. Used tool for extracting abstracts leads concatenate words of section headings with the first word of the section. For instance, we observe words such as ConclusionHigher and ConclusionsRT etc. The detection and identification of such words is done by sampling of medicine-related publications with human intervention. Detected concatenate words are split into two words. For instance, the word ‘ConclusionHigher’ is split into ‘Conclusion’ and ‘Higher’.The section headings in such abstracts are listed below:

Background Method(s) Design Theoretical Measurement(s) Location Aim(s) Methodology Process Abstract Population Approach Objective(s) Purpose(s) Subject(s) Introduction Implication(s) Patient(s) Procedure(s) Hypothesis Measure(s) Setting(s) Limitation(s) Discussion Conclusion(s) Result(s) Finding(s) Material (s) Rationale(s) Implications for health and nursing policyStep 5: Extracting (Sub-setting) the Data Based on Lengths of AbstractsAfter correction, the lengths of abstracts are calculated. ‘Length’ indicates the total number of words in the text, calculated by the same rule as for Microsoft Word ‘word count’ [5].According to APA style manual [6], an abstract should contain between 150 to 250 words. In LSC, we decided to limit length of abstracts from 30 to 500 words in order to study documents with abstracts of typical length ranges and to avoid the effect of the length to the analysis.

Step 6: [Version 2] Cleaning Copyright Notices, Permission polices, Journal Names and Conference Names from LSC Abstracts in Version 1Publications can include a footer of copyright notice, permission policy, journal name, licence, author’s right or conference name below the text of abstract by conferences and journals. Used tool for extracting and processing abstracts in WoS database leads to attached such footers to the text. For example, our casual observation yields that copyright notices such as ‘Published by Elsevier ltd.’ is placed in many texts. To avoid abnormal appearances of words in further analysis of words such as bias in frequency calculation, we performed a cleaning procedure on such sentences and phrases in abstracts of LSC version 1. We removed copyright notices, names of conferences, names of journals, authors’ rights, licenses and permission policies identified by sampling of abstracts.Step 7: [Version 2] Re-extracting (Sub-setting) the Data Based on Lengths of AbstractsThe cleaning procedure described in previous step leaded to some abstracts having less than our minimum length criteria (30 words). 474 texts were removed.Step 8: Saving the Dataset into CSV FormatDocuments are saved into 34 CSV files. In CSV files, the information is organised with one record on each line and parts of abstract, title, list of authors, list of categories, list of research areas, and times cited is recorded in fields.To access the LSC for research purposes, please email to ns433@le.ac.uk.References[1]Web of Science. (15 July). Available: https://apps.webofknowledge.com/ [2]WoS Subject Categories. Available: https://images.webofknowledge.com/WOKRS56B5/help/WOS/hp_subject_category_terms_tasca.html [3]Research Areas in WoS. Available: https://images.webofknowledge.com/images/help/WOS/hp_research_areas_easca.html [4]Times Cited in WoS Core Collection. (15 July). Available: https://support.clarivate.com/ScientificandAcademicResearch/s/article/Web-of-Science-Times-Cited-accessibility-and-variation?language=en_US [5]Word Count. Available: https://support.office.com/en-us/article/show-word-count-3c9e6a11-a04d-43b4-977c-563a0e0d5da3 [6]A. P. Association, Publication manual. American Psychological Association Washington, DC, 1983.

DATA MINING THE GALAXY ZOO MERGERS STEVEN BAEHR, ARUN VEDACHALAM, KIRK BORNE, AND DANIEL SPONSELLER Abstract. Collisions between pairs of galaxies usually end in the coalescence (merger) of the two galaxies. Collisions and mergers are rare phenomena, yet they may signal the ultimate fate of most galaxies, including our own Milky Way. With the onset of massive collection of astronomical data, a computerized and automated method will be necessary for identifying those colliding galaxies worthy of more detailed study. This project researches methods to accomplish that goal. Astronomical data from the Sloan Digital Sky Survey (SDSS) and human-provided classifications on merger status from the Galaxy Zoo project are combined and processed with machine learning algorithms. The goal is to determine indicators of merger status based solely on discovering those automated pipeline-generated attributes in the astronomical database that correlate most strongly with the patterns identified through visual inspection by the Galaxy Zoo volunteers. In the end, we aim to provide a new and improved automated procedure for classification of collisions and mergers in future petascale astronomical sky surveys. Both information gain analysis (via the C4.5 decision tree algorithm) and cluster analysis (via the Davies-Bouldin Index) are explored as techniques for finding the strongest correlations between human-identified patterns and existing database attributes. Galaxy attributes measured in the SDSS green waveband images are found to represent the most influential of the attributes for correct classification of collisions and mergers. Only a nominal information gain is noted in this research, however, there is a clear indication of which attributes contribute so that a direction for further study is apparent.

We present a ProgSnap2-based dataset containing anonymized logs of over 34,000 programming events exhibited by 81 programming students in Scratch, a visual programming environment, during our designed study as described in the paper "Semi-Automatically Mining Students' Common Scratch Programming Behaviors." We also include a list of approx. 3100 mined sequential patterns of programming processes that are performed by at least 10% of the 62 of the 81 students who are novice programmers, and represent maximal patterns generated by the MG-FSM algorithm while allowing a gap of one programming event. README.txt — overview of the dataset and its propertiesmainTable.csv — main event table of the dataset holding rows of programming eventscodeState.csv — table holding XML representations of code snapshots at the time of each programming eventdatasetMetadata.csv — describes features of the datasetScratch-SeqPatterns.txt — list of sequential patterns mined from the Main Event Table

The complete dataset used in the analysis comprises 36 samples, each described by 11 numeric features and 1 target. The attributes considered were caspase 3/7 activity, Mitotracker red CMXRos area and intensity (3 h and 24 h incubations with both compounds), Mitosox oxidation (3 h incubation with the referred compounds) and oxidation rate, DCFDA fluorescence (3 h and 24 h incubations with either compound) and oxidation rate, and DQ BSA hydrolysis. The target of each instance corresponds to one of the 9 possible classes (4 samples per class): Control, 6.25, 12.5, 25 and 50 µM for 6-OHDA and 0.03, 0.06, 0.125 and 0.25 µM for rotenone. The dataset is balanced, it does not contain any missing values and data was standardized across features. The small number of samples prevented a full and strong statistical analysis of the results. Nevertheless, it allowed the identification of relevant hidden patterns and trends.

Exploratory data analysis, information gain, hierarchical clustering, and supervised predictive modeling were performed using Orange Data Mining version 3.25.1 [41]. Hierarchical clustering was performed using the Euclidean distance metric and weighted linkage. Cluster maps were plotted to relate the features with higher mutual information (in rows) with instances (in columns), with the color of each cell representing the normalized level of a particular feature in a specific instance. The information is grouped both in rows and in columns by a two-way hierarchical clustering method using the Euclidean distances and average linkage. Stratified cross-validation was used to train the supervised decision tree. A set of preliminary empirical experiments were performed to choose the best parameters for each algorithm, and we verified that, within moderate variations, there were no significant changes in the outcome. The following settings were adopted for the decision tree algorithm: minimum number of samples in leaves: 2; minimum number of samples required to split an internal node: 5; stop splitting when majority reaches: 95%; criterion: gain ratio. The performance of the supervised model was assessed using accuracy, precision, recall, F-measure and area under the ROC curve (AUC) metrics.

ABSTRACT To enhance speed and agility in interpreting physiological quality tests of seeds, The use of algorithms has emerged. This study aimed to identify suitable machine learning models to assist in the precise management of seed lot quality. Soybean lots from two companies were assessed using the Supplied Test Set, Cross-Validation (with 8, 10, and 12 folds), and Percentage Split (with 66% and 70%) methods. Variables analyzed through Tetrazolium tests included vigor, viability, mechanical damage, moisture damage, bed bug damage, and water content. Method performance was determined by Kappa, Precision, and ROC Area metrics. Classification Via Regression and J48 algorithms were employed. The technique utilizing 66% of data for training achieved 93.55% accuracy, with Precision and ROC Area reaching 94.50% for the J48 algorithm. Applying the cross-validation method with 10 folds resulted in 90.22% of correctly classified instances, with a ROC Area outcome like the previous method. Tetrazolium Vigor was the primary attribute used. However, these results are specific to this study's database, and careful planning is necessary to select the most effective application methods.

We put forward a new approach to studying issue definition within the context of policy diffusion. Most studies of policy diffusion---which is the process by which policymaking in one government affects policymaking in other governments---have focused on policy adoptions. We shift the focus to an important but neglected aspect of this process: the issue-definition stage. We use topic models to estimate how policies are framed during this stage and how these frames are predicted by prior policy adoptions. Focusing on smoking restriction in U.S. states, our analysis draws upon an original dataset of over 52,000 paragraphs from newspapers covering 49 states between 1996 and 2013. We find that frames regarding the policy's concrete implications are predicted by prior adoptions in other states, while frames regarding its normative justifications are not. Our approach and findings open the way for a new perspective to studying policy diffusion in many different areas.

As semiconductor devices are miniaturized, the importance of atomic layer deposition (ALD) technology is growing. When designing ALD precursors, it is important to consider the melting point, because the precursors should have melting points lower than the process temperature. However, obtaining melting point data is challenging due to experimental sensitivity and high computational costs. As a result, a comprehensive and well-organized database for the melting point of the OMCs has not been fully reported yet. Therefore, in this study, we constructed a database of melting points for 1,845 OMCs, including 58 metal and 6 metalloid elements. The database contains CAS numbers, molecular formulas, and structural information and was constructed through automatic extraction and systematic curation. The melting point information was extracted using two methods: 1) 1,434 materials from 11 chemical vendor databases and 2) 411 materials identified through natural language processing (NLP) techniques with an accuracy of 86.3%, based on 2,096 scientific papers published over the past 29 years. In our database, the OMCs contain up to around 250 atoms and have melting points that range from −170 to 1610 °C. The main source is the Chemsrc database, accounting for 607 materials (32.9%), and Fe is the most common central metal or metalloid element (15.0%), followed by Si (11.6%) and B (6.7%). To validate the utilization of the constructed database, a multimodal neural network model was developed integrating graph-based and feature-based information as descriptors to predict the melting points of the OMCs but moderate performance. We believe the current approach reduces the time and cost associated with hand-operated data collection and processing, contributing to effective screening of potentially promising ALD precursors and providing crucial information for the advancement of the semiconductor industry.

This folder contains the synthetically generated dataset (process model and event logs) containing process data of a synthetically designed package delivery process, as described in [1]. The event logs present simulations of a process model, each with an incorporated issue, be it a behavioral deviation, i.e., where the process is differently exhibited with regard to the expected behavior described by the process model, or a recording error, i.e., where the execution of the process is recorded differently with regard to how it is exhibited. Each issue is added to the process model through a model transformation providing ground truth to the discrepancies introduced in the simulated event log.

The package delivery process starts with the choice of home or depot delivery, after which the package queues for a warehouse employee to pick and load it into a van. In case of home delivery, a courier drives off and rings a door after which he continues to either immediately hand over the package, or deliver it at the corresponding depot after registration, where it is left for collection. Alternatively, for depot delivery, "ringing" and therefore also "deliver at home" is omitted in the subprocess.

models/delivery_base_model.json contains the specification of the process model that incorporates this "expected behavior", and is depicted in models/delivery_base_model.pdf.

On top of this, six patterns of behavioral deviations (BI) and six patterns of recording errors (RI) are applied to the base model:

BI5: Overtaking in the FIFO queue for picking packages;

BI7: Switching roles from a courier to that of a warehouse employee;

BI10: Batching is ignored, leaving with a delivery van before it was fully loaded;

BI3: Skipping the activity of ringing, modeling behavior where e.g., the door was already opened upon arrival;

BI9: Different resource memory where the package is delivered to a different depot than where it is registered;

BI2: Multitasking of couriers during the delivery of multiple packages, modeling interruption of a delivery;

RI1: Incorrect event, recording an order for depot delivery when it was intended for home delivery;

RI2: Incorrect event, vice versa, i.e., recording an order for home delivery when it was intended for depot delivery;

RI3: Missing event for the activity of loading a package in a truck;

RI4: Missing object of the involved van for loading, e.g., due to a temporary connection failure of a recording device;

RI5: Incorrect object of the involved courier when ringing, e.g., due to not logging out by the courier on the previous shift;

RI6: Missing positions for the recording of the delivery and the collection at a depot, e.g., due to coarse timestamp logging.

The behavior of each deviation pattern is added separately to the base model, resulting in twelve process models, accordingly named models/package_delivery_

Each model is simulated resulting in twelve logs, accordingly named logs/package_delivery_

All models and corresponding generated logs with the applied patterns are also available at gitlab.com/dominiquesommers/mira/-/tree/main/mira/simulation, which additionally includes scripts to load and process the data.

We refer to [1] for more information on the dataset.

[1] Dominique Sommers, Natalia Sidorova, Boudewijn F. van Dongen. A ground truth approach for assessing process mining techniques. arXiv preprint, https://doi.org/10.48550/arXiv.2501.14345, 2025.

This dataset contains results of the experiment to analyze information preservation and recovery by different event log abstractions in process mining described in: Sander J.J. Leemans, Dirk Fahland "Information-Preserving Abstractions of Event Data in Process Mining"
Knowledge and Information Systems, ISSN: 0219-1377 (Print) 0219-3116 (Online), accepted May 2019

The experiment results were obtained with: https://doi.org/10.5281/zenodo.3243981

This dataset contains a variety of publicly available real-life event logs. We derived two types of Petri nets for each event log with two state-of-the-art process miners : Inductive Miner (IM) and Split Miner (SM). Each event log-Petri net pair is intended for evaluating the scalability of existing conformance checking techniques.We used this data-set to evaluate the scalability of the S-Component approach for measuring fitness. The dataset contains tables of descriptive statistics of both process models and event logs. In addition, this dataset includes the results in terms of time performance measured in milliseconds for several approaches for both multi-threaded and single-threaded executions. Last, the dataset contains a cost-comparison of different approaches and reports on the degree of over-approximation of the S-Components approach. The description of the compared conformance checking techniques can be found here: https://arxiv.org/abs/1910.09767. Update:The dataset has been extended with the event logs of the BPIC18 and BPIC19 logs. BPIC19 is actually a collection of four different processes and thus was split into four event logs. For each of the additional five event logs, again, two process models have been mined with inductive and split miner. We used the extended dataset to test the scalability of our tandem repeats approach for measuring fitness. The dataset now contains updated tables of log and model statistics as well as tables of the conducted experiments measuring execution time and raw fitness cost of various fitness approaches. The description of the compared conformance checking techniques can be found here: https://arxiv.org/abs/2004.01781.Update: The dataset has also been used to measure the scalability of a new Generalization measure based on concurrent and repetitive patterns. : A concurrency oracle is used in tandem with partial orders to identify concurrent patterns in the log that are tested against parallel blocks in the process model. Tandem repeats are used with various trace reduction and extensions to define repetitive patterns in the log that are tested against loops in the process model. Each pattern is assigned a partial fulfillment. The generalization is then the average of pattern fulfillments weighted by the trace counts for which the patterns have been observed. The dataset no includes the time results and a breakdown of Generalization values for the dataset.

Technical notes and documentation on the common data model of the project CONCEPT-DM2.

This publication corresponds to the Common Data Model (CDM) specification of the CONCEPT-DM2 project for the implementation of a federated network analysis of the healthcare pathway of type 2 diabetes.

Aims of the CONCEPT-DM2 project:

General aim: To analyse chronic care effectiveness and efficiency of care pathways in diabetes, assuming the relevance of care pathways as independent factors of health outcomes using data from real life world (RWD) from five Spanish Regional Health Systems.

Main specific aims:

• To characterize the care pathways in patients with diabetes through the whole care system in terms of process indicators and pharmacologic recommendations
• To compare these observed care pathways with the theoretical clinical pathways derived from the clinical practice guidelines
• To assess if the adherence to clinical guidelines influence on important health outcomes, such as cardiovascular hospitalizations.
• To compare the traditional analytical methods with process mining methods in terms of modeling quality, prediction performance and information provided.

Study Design: It is a population-based retrospective observational study centered on all T2D patients diagnosed in five Regional Health Services within the Spanish National Health Service. We will include all the contacts of these patients with the health services using the electronic medical record systems including Primary Care data, Specialized Care data, Hospitalizations, Urgent Care data, Pharmacy Claims, and also other registers such as the mortality and the population register.

Cohort definition: All patients with code of Type 2 Diabetes in the clinical health records

• Inclusion criteria: patients that, at 01/01/2017 or during the follow-up from 01/01/2017 to 31/12/2022 had active health card (active TIS - tarjeta sanitaria activa) and code of type 2 diabetes (T2D, DM2 in spanish) in the clinical records of primary care (CIAP2 T90 in case of using CIAP code system)
• Exclusion criteria:
  • patients with no contact with the health system from 01/01/2017 to 31/12/2022
  • patients that had a T1D (DM1) code opened after the T2D code during the follow-up.
• Study period. From 01/01/2017 to 31/12/2022

Files included in this publication:

• Datamodel_CONCEPT_DM2_diagram.png
• Common data model specification (Datamodel_CONCEPT_DM2_v.0.1.0.xlsx)
• Synthetic datasets (Datamodel_CONCEPT_DM2_sample_data)
  • sample_data1_dm_patient.csv
  • sample_data2_dm_param.csv
  • sample_data3_dm_patient.csv
  • sample_data4_dm_param.csv
  • sample_data5_dm_patient.csv
  • sample_data6_dm_param.csv
  • sample_data7_dm_param.csv
  • sample_data8_dm_param.csv
• Datamodel_CONCEPT_DM2_explanation.pptx

Anomaly detection is a process of identifying items, events or observations, which do not conform to an expected pattern in a dataset or time series. Current and future missions and our research communities challenge us to rapidly identify features and anomalies in complex and voluminous observations to further science and improve decision support. Given this data intensive reality, we propose to develop an anomaly detection system, called OceanXtremes, powered by an intelligent, elastic Cloud-based analytic service backend that enables execution of domain-specific, multi-scale anomaly and feature detection algorithms across the entire archive of ocean science datasets. A parallel analytics engine will be developed as the key computational and data-mining core of OceanXtreams' backend processing. This analytic engine will demonstrate three new technology ideas to provide rapid turn around on climatology computation and anomaly detection: 1. An adaption of the Hadoop/MapReduce framework for parallel data mining of science datasets, typically large 3 or 4 dimensional arrays packaged in NetCDF and HDF. 2. An algorithm profiling service to efficiently and cost-effectively scale up hybrid Cloud computing resources based on the needs of scheduled jobs (CPU, memory, network, and bursting from a private Cloud computing cluster to public cloud provider like Amazon Cloud services). 3. An extension to industry-standard search solutions (OpenSearch and Faceted search) to provide support for shared discovery and exploration of ocean phenomena and anomalies, along with unexpected correlations between key measured variables. We will use a hybrid Cloud compute cluster (private Eucalyptus on-premise at JPL with bursting to Amazon Web Services) as the operational backend. The key idea is that the parallel data-mining operations will be run 'near' the ocean data archives (a local 'network' hop) so that we can efficiently access the thousands of (say, daily) files making up a three decade time-series, and then cache key variables and pre-computed climatologies in a high-performance parallel database. OceanXtremes will be equipped with both web portal and web service interfaces for users and applications/systems to register and retrieve oceanographic anomalies data. By leveraging technology such as Datacasting (Bingham, et.al, 2007), users can also subscribe to anomaly or 'event' types of their interest and have newly computed anomaly metrics and other information delivered to them by metadata feeds packaged in standard Rich Site Summary (RSS) format. Upon receiving new feed entries, users can examine the metrics and download relevant variables, by simply clicking on a link, to begin further analyzing the event. The OceanXtremes web portal will allow users to define their own anomaly or feature types where continuous backend processing will be scheduled to populate the new user-defined anomaly type by executing the chosen data mining algorithm (i.e. differences from climatology or gradients above a specified threshold). Metadata on the identified anomalies will be cataloged including temporal and geospatial profiles, key physical metrics, related observational artifacts and other relevant metadata to facilitate discovery, extraction, and visualization. Products created by the anomaly detection algorithm will be made explorable and subsettable using Webification (Huang, et.al, 2014) and OPeNDAP (http://opendap.org) technologies. Using this platform scientists can efficiently search for anomalies or ocean phenomena, compute data metrics for events or over time-series of ocean variables, and efficiently find and access all of the data relevant to their study (and then download only that data).

Full title: Mining Distance-Based Outliers in Near Linear Time with Randomization and a Simple Pruning Rule Abstract: Defining outliers by their distance to neighboring examples is a popular approach to finding unusual examples in a data set. Recently, much work has been conducted with the goal of finding fast algorithms for this task. We show that a simple nested loop algorithm that in the worst case is quadratic can give near linear time performance when the data is in random order and a simple pruning rule is used. We test our algorithm on real high-dimensional data sets with millions of examples and show that the near linear scaling holds over several orders of magnitude. Our average case analysis suggests that much of the efficiency is because the time to process non-outliers, which are the majority of examples, does not depend on the size of the data set.

The Process Discovery approach described in the submitted document is directed towards discovery of process models from a Training Event log representing 10 different real time business process executions, and cross-validating the derived model with a set of two Test Event logs provided for evaluation of the process discovery technique. Each of the Test event logs ((test_log_april_1 to test_log_april_10) and (test_log_may_1 to test_log_may_10)) represents part of the model from the Training Log with complete total of 20 traces for each of the logs, and are characterized by having 10 traces that can be replayed (allowed) and 10 traces that cannot be replayed (disallowed) by the model. The total number of traces for the Test event logs (i.e. April log and May log) is therefore ((10 logs x 20 traces) x 2) = 400 Traces. Our aim is to carry out a classification task to determine the 400 individual traces that makes up the two test event log and then provide a Petri Net representation of the Training model as well as Business Process Model Notation (BPMN) mapping that allows for testing and evaluation of the behaviours/traces recorded in the Test logs. The objective of the proposed approach is to discover and provide process models that matches the original process models in term of balancing between “overfitting” and “underfitting”. A process model is seen as overfitting (the event log) if it is too restrictive, disallowing behaviour which is part of the underlying process. On the other hand, it is underfitting (the reality) if it is not restrictive enough, allowing behaviour which is not part of the underlying process. Following this challenge, we aim to provide a model which is as good in balancing “overfitting” and “underfitting” as it is able to correctly classify the traces that can be replayed in the “test” event log: Thus, • Given a trace (t) representing real process behaviour, the process model (m) classifies it as allowed, or • Given a trace (t) representing a behaviour not related to the process, the process model (m) classifies it as disallowed. The submitted document contains the classification attempts for the events logs provided and discusses the replaying semantics of the process modelling notation that has been employed. In other words, we discuss how, given any process trace t (for the Test event Log) and process model m (for the training log) in the discovered Petri Net and BPMN replaying notation, it can be unambiguously determined whether or not trace t can be replayed on model (m). We also provide a description of the tools used to discover the process models as well as checking the result of the classification task.