{"id":5626,"date":"2020-04-23T00:15:00","date_gmt":"2020-04-23T00:15:00","guid":{"rendered":"https:\/\/www.dab-europe.com\/?post_type=articles&#038;p=5626"},"modified":"2024-07-30T13:36:30","modified_gmt":"2024-07-30T13:36:30","slug":"the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2","status":"publish","type":"articles","link":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/","title":{"rendered":"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\"><span class=\"size-3\">Part 2 \u2013 \u201eCluster\u201c in \u201eACL\u2122 Robotics\u201c<\/span><\/h2>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">This is the second part of a short series of blog posts where we present you two machine learning procedures with the analytic&nbsp;<a href=\"https:\/\/old.dab-group.com\/en-US\/diligent-software\/robotics-enterprise\"><strong>Software ACL Robotics.<\/strong><\/a>&nbsp;The \u201cACL\u2122 Robotics\u201d software solution, which has been established on the market for many years, supports the manual and automated analysis of large amounts of data. In addition to a multitude of interfaces to SAP&nbsp;<a href=\"https:\/\/www.dab-europe.com\/en\/solutions\/sap-connector\/\"><strong>(via \u201cSAP Connector\u201d)<\/strong>,<\/a>&nbsp;Salesforce, Google Hive, Amazon Redshift, Outlook, PDF imports or any ODBC data source, a script language helps to automate analysis steps. The software developer Galvanize allocates this to the field of<a href=\"https:\/\/www.bctechnology.com\/news\/2019\/1\/8\/ACL-Breaks-Ground-with-Robotic-Process-Automation-for-Governance-Risk-&amp;-Compliance.cfm\" target=\"_blank\" rel=\"noreferrer noopener\">&nbsp;<strong>RPA (Robotic Process Automation)<\/strong>.<\/a>&nbsp;Individual analytic steps are performed by several analytic commands, such as sorting, summarizing, joining and relating, to name only a few. With the Version 14 these analytic commands have been extended by three machine learning commands named as \u201cTrain\u201d, \u201cPredict\u201d and \u201cCluster\u201d. In these two blogposts we will introduce these three commands to you by taking examples from out of the everyday business. For all ACL users, we also offer the opportunity to download ACL projects, allowing you to try out each command, step by step.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This blog post is, due to its length, divided into two parts:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/www.dab-europe.com\/en\/articles\/die-machine-learning-befehle-train-predict-und-cluster-am-praxisbeispiel-erklart\/\"><strong>Part 1<\/strong><\/a>&nbsp;deals with the \u201cTrain\u201d and \u201cPredict\u201d commands<\/li>\n\n\n\n<li><strong>Part 2 deals with the \u201cCluster\u201d command, which is based on the k-means algorithm<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">The example for Part 1 was taken from the field of sales: Customers order goods of various different values. Of course, as part of the day-to-day business, this may involve returns. Customers return goods for various reasons and usually receive a refund. In the second part, we are taking up the challenge of dividing customers into groups with similar behaviour. As our example is a B2B example, our customers are likewise companies. As payment term we have agreed on 30 days net, the agreement is stored in the customer master data. Let us take, for example, the following question:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>If you grant your customers a payment term of 30 days, are there groups of companies which behave similar towards the deadline?<\/strong><\/p>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">We will answer this question by using the \u201cCluster\u201d command in ACL. In Figure 7 you can see a notional dataset of one year plotted. Each point in the graph represents a customer. The annual turnover is plotted on the y-axis. The average payment delay in days is plotted on the x-axis, these values are always rounded. The average always emerges from several payment delays in the reviewed year. An average payment delay of greater than zero means that the invoices of the company observed are, on average, settled prior to the expiry of the 30-day payment period. A value of less than zero means that the respective invoices of the company observed are, on average, paid after the expiry of the 30-day payment period. Companies on the vertical blue line paid, on average, on time, on the last possible day.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter is-resized\"><img decoding=\"async\" src=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/Publ\/cD\/Abbildung_Clustering_1.png\" alt=\"\" style=\"width:394px;height:auto\"\/><figcaption class=\"wp-element-caption\"><sub>Figure 7: Average payment delay in days and annual turnover of 1,449 customers<\/sub><\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">We are, with the aid of ACL and the \u201cCluster\u201d command, grouping below the data from Figure 7, in order to answer the question posed at the outset. The \u201cCluster\u201d function is based on the&nbsp;<a href=\"https:\/\/help.highbond.com\/helpdocs\/analytics\/141\/user-guide\/en-us\/Content\/analyzing_data\/summarizing_data\/clustering_data.htm\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>k-Means algorithm<\/strong>.<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><b><span class=\"background-color-blue-1 color-white\">Workshop:<\/span><\/b><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Downloading data:&nbsp;Open ACL and the project \u201ccluster_customers.acl\u201d. This contains the following two tables: \u201cturnover_and_delay\u201d and \u201cturnover_and_delay_with_GAP\u201d. The clustering is carried out below based on the first table. The second table does not, for now, come into the picture.<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Clustering data:&nbsp;Click on \u201cturnover_and_delay\u201d in the side bar. You will now see the associated table in the Basic View. Select \u201cMachine Learning\u201d -&gt; \u201cCluster\u201d. Next, change the parameters and settings, and then initiate the clustering operation by pressing \u201cOK\u201d (cf. Figure 8).<\/li>\n<\/ul>\n\n\n\n<ol class=\"wp-block-list\">\n<li><\/li>\n<\/ol>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"848\" height=\"692\" src=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d.png\" alt=\"\" class=\"wp-image-5715\" style=\"width:676px;height:auto\" srcset=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d.png 848w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d-300x245.png 300w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d-150x122.png 150w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d-768x627.png 768w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d-772x630.png 772w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d-437x357.png 437w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_screenshot4_4fd2a4958d-320x261.png 320w\" sizes=\"(max-width: 848px) 100vw, 848px\" \/><figcaption class=\"wp-element-caption\"><sub>Figure 8: Input mask for clustering with altered parameters<\/sub><div id=\"c5196\" class=\"page-padding-left-right--narrow \" style=\"box-sizing: border-box; padding-left: 1.75rem; padding-right: 1.75rem; color: rgb(60, 60, 60); font-family: Roboto, sans-serif; font-size: 16px; text-align: start; white-space-collapse: collapse;\"><div class=\"grid   \" style=\"box-sizing: border-box; list-style: none; margin: 0px 0px -3.1875rem -3.1875rem; padding: 0px;\"><div class=\"grid__item one-whole palm--one-whole\" style=\"box-sizing: border-box; display: inline-block; padding-left: 3.1875rem; margin-bottom: 3.1875rem; vertical-align: top; width: 1007px;\"><ul style=\"padding: 0px 0px 0px 1.75em; margin: 0px 0px 1em;\"><\/ul><\/div><\/div><\/div><\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>&#8220;Number of clusters (K value)\u201d: This value determines into how many clusters the dataset being inspected is divided. ACL&nbsp;<a href=\"https:\/\/help.highbond.com\/helpdocs\/analytics\/141\/user-guide\/en-us\/Content\/analyzing_data\/summarizing_data\/clustering_data.htm\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>recommends<\/strong>&nbsp;<\/a>trying out various values. In addition, it is advised to make a first attempt using 8 &#8211; 10 clusters. With the aid of Figure 7, we have, in this example, chosen to do 4 clusters.<\/li>\n\n\n\n<li>\u201cNumber of initializations\u201d: K-Means depends upon the starting points selected. For this reason, the algorithm is automatically repeated several times over, using various different starting points, in the ACL software. The best run, i.e. the best clustering, is subsequently selected. \u201cNumber of initializations\u201d lets you determine how often the algorithm is supposed to be executed using various different starting points..<\/li>\n\n\n\n<li>\u201cMaximum number of iterations\u201d: How many iterations should the clustering algorithm carry out per initialization, as a maximum. The higher this value is set, the more suitable are the clusterings resulting from each run.<\/li>\n\n\n\n<li>\u201cSeed\u201d: K-Means depends upon the randomly selected starting points. \u201cSeed\u201d influences the random number generator. Should you again and again repeat the clustering using the same value for \u201cSeed\u201d and the identical parameters, you will get the same result every time. Should you only change \u201cSeed\u201d, and repeat the clustering, you will get somewhat different clusters each time. This is based on the choice of randomly selected starting values that depend upon \u201cSeed\u201d.<\/li>\n\n\n\n<li>\u201cPre-processing\u201d: Here you can edit your fields prior to clustering. This makes sense if your key fields show extreme differences in the value ranges.<\/li>\n\n\n\n<li>\u201cCluster on\u2026\u201d: Here you can choose the fields based on which the clusters should be formed. In our case, the average payment delay.<\/li>\n\n\n\n<li>\u201cOther Fields\u2026\u201d: Here you can specify fields which should be displayed in the results table alongside the aforementioned fields.<\/li>\n\n\n\n<li>\u201cIf\u2026\u201d and \u201cMore\u201d: Here you can optionally exclude entries from the dataset.<\/li>\n\n\n\n<li>\u201cTo\u2026\u201d: Here you can specify the name of the results table.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Once the calculation has been completed, you will see the \u201cclustered_customers\u201d table in your side bar. The first k lines in this table always contain the mid-points of the clusters found. The field \u201cCluster\u201d specifies the associated cluster for each record. The field \u201cDistance\u201d contains the distance to the associated mid-point of the cluster. In Figure 9, the clustering from within the ACL software is visualised by means of colour.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"500\" height=\"643\" src=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee.png\" alt=\"\" class=\"wp-image-5717\" style=\"width:434px;height:auto\" srcset=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee.png 500w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee-233x300.png 233w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee-117x150.png 117w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee-490x630.png 490w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee-278x357.png 278w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_2_f247c7d5ee-203x261.png 203w\" sizes=\"(max-width: 500px) 100vw, 500px\" \/><figcaption class=\"wp-element-caption\"><sub>Figure 9: Average payment delay in days and annual turnover of 1,449 customers, clustered into four groups<\/sub><\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">The clustering generates the following groups:<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>Cluster colour<\/th><th><strong>Average payment delay MIN<\/strong><\/th><th><strong>Average payment delay MAX<\/strong><\/th><\/tr><\/thead><tbody><tr><td>Blue<\/td><td>-88<\/td><td>-38<\/td><\/tr><tr><td>Green<\/td><td>-33<\/td><td>-9<\/td><\/tr><tr><td>Black<\/td><td>-8<\/td><td>10<\/td><\/tr><tr><td>Red<\/td><td>11<\/td><td>29<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">The blue cluster contains 1.8% of the companies which have paid late in the year observed, on average between 88 and 38 days. For the green cluster (3.8%), the interpretation is made analogously, using the values from the above table. The black cluster contains customers (34.6%) who, on average, paid their invoices on time or not. The red cluster (59.8%) contains exclusively companies which paid on average on time. When looking at Figure 9, it is noticeable that k-Means divides the data into groups relatively well. The clustering nevertheless has a decisive weak point. It would be desirable for every group to be located either fully left or fully right of zero. Interpreting the result is easier if the clusters only contain companies that paid either on average on time or on average not on time. The black cluster is, however, located on both the left and the right side of the zero. For this reason, we use a trick and repeat the above procedure. We generate an artificial gap around zero (cf. Figure 10). All points which are located in the area between -8 inclusive and 0 exclusive, in relation to the x-axis, are moved to exactly x=-8. Points which are located between 0 inclusive and 8 inclusive are moved to exactly x=8.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"500\" height=\"644\" src=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e.png\" alt=\"\" class=\"wp-image-5719\" srcset=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e.png 500w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e-233x300.png 233w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e-116x150.png 116w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e-489x630.png 489w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e-277x357.png 277w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_3_eb3a36ba4e-203x261.png 203w\" sizes=\"(max-width: 500px) 100vw, 500px\" \/><figcaption class=\"wp-element-caption\"><sub>Figure 10: Transformed average payment delay in days and annual turnover of 1,449 customers<\/sub><\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">We then repeat the clustering in ACL. The resulting grouping can be seen from Figure 11.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"500\" height=\"644\" src=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef.png\" alt=\"\" class=\"wp-image-5721\" srcset=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef.png 500w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef-233x300.png 233w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef-116x150.png 116w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef-489x630.png 489w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef-277x357.png 277w, https:\/\/www.dab-europe.com\/wp-content\/uploads\/2020\/04\/csm_Abbildung_Clustering_4_1a733d4cef-203x261.png 203w\" sizes=\"(max-width: 500px) 100vw, 500px\" \/><figcaption class=\"wp-element-caption\"><sub>Figure 11: Transformed average payment delay in days and annual turnover of 1,449 customers, clustered into four groups<\/sub><\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">Using this trick, every cluster is, as desired, located either fully left or fully right of zero. The procedure also works with constants other than -8 and 8. The grouping has been determined by ACL to be as follows:<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>Cluster colour<\/th><th>Average payment delay MIN<\/th><th>Average payment delay MAX<\/th><\/tr><\/thead><tbody><tr><td>Blue<\/td><td>-88<\/td><td>-31<\/td><\/tr><tr><td>Green<\/td><td>-29<\/td><td>-1<\/td><\/tr><tr><td>Black<\/td><td>0<\/td><td>14<\/td><\/tr><tr><td>Red<\/td><td>15<\/td><td>29<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">The grouping can be interpreted as follows: All 1,449 data points can be divided into four sub-sets. On the one hand, 88% of customers, them at the right side of zero, paid on average on time. Whereas, on the other hand, 12% are located to the left of zero &#8211; these companies did not, on average, pay on time. On the right-hand side of the zero, the customers were divided into two further groups: Into the black cluster, which contains customers which, on average, paid between 14 days prior to the deadline and the last possible date. The red cluster contains companies which, on average, already paid 29 to 15 days prior to the deadline. The interpretation is made analogously for the green and blue groupings, only paying attention to the changed algebraic sign.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The procedure presented involving the \u201cCluster\u201d command from ACL belongs to the Unsupervised Learning. In comparison to the example with the estimated return value from Part 1, no value is estimated, but the data is grouped. Thus, no \u201cfeature variable\u201d exists.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In this second part of our blog post introducing Machine Learning, we have shown you how you can group your data. The necessary steps you need to perform can be summarised as follows:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Create a table with the desired \u201ckey variables&#8221;, based on which k-Means is supposed to cluster.<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Carry out the \u201cCluster\u201d function with suitable parameters<\/li>\n<\/ul>\n\n\n\n<ol class=\"wp-block-list\">\n<li><\/li>\n<\/ol>\n\n\n\n<p class=\"wp-block-paragraph\">By selecting the \u201ckey variables\u201d, as well as said parameters, you can significantly influence the clustering.<\/p>\n\n\n\n<p class=\"align-justify wp-block-paragraph\">We hope you enjoyed reading this blogpost regarding Machine Learning and would be happy, if it can support you, when considering on how to use Machine Learning in your everyday business. Have fun trying out these methods and please do not hesitate to contact us at any time, if you have any questions.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Part 2 \u2013 \u201eCluster\u201c in \u201eACL\u2122 Robotics\u201c This is the second part of a short series of blog posts where we present you two machine learning procedures with the analytic&nbsp;Software ACL Robotics.&nbsp;The \u201cACL\u2122 Robotics\u201d software solution, which has been established on the market for many years, supports the manual and automated analysis of large amounts [&hellip;]<\/p>\n","protected":false},"author":17,"featured_media":5585,"template":"","articles_category":[],"articles_tag":[92,108],"class_list":["post-5626","articles","type-articles","status-publish","has-post-thumbnail","hentry","articles_tag-sap","articles_tag-ki"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2 - dab<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2 - dab\" \/>\n<meta property=\"og:description\" content=\"Part 2 \u2013 \u201eCluster\u201c in \u201eACL\u2122 Robotics\u201c This is the second part of a short series of blog posts where we present you two machine learning procedures with the analytic&nbsp;Software ACL Robotics.&nbsp;The \u201cACL\u2122 Robotics\u201d software solution, which has been established on the market for many years, supports the manual and automated analysis of large amounts [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/\" \/>\n<meta property=\"og:site_name\" content=\"dab\" \/>\n<meta property=\"article:modified_time\" content=\"2024-07-30T13:36:30+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2024\/07\/G19A0698-1024x683.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"683\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"11 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/\",\"url\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/\",\"name\":\"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2 - dab\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.dab-europe.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/G19A0698-scaled.jpg\",\"datePublished\":\"2020-04-23T00:15:00+00:00\",\"dateModified\":\"2024-07-30T13:36:30+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/#primaryimage\",\"url\":\"https:\\\/\\\/www.dab-europe.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/G19A0698-scaled.jpg\",\"contentUrl\":\"https:\\\/\\\/www.dab-europe.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/G19A0698-scaled.jpg\",\"width\":2560,\"height\":1706},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Articles\",\"item\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/articles\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/#website\",\"url\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/\",\"name\":\"dab\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/#organization\",\"name\":\"dab\",\"url\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/www.dab-europe.com\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/dab_logo-main-color-scaled.png\",\"contentUrl\":\"https:\\\/\\\/www.dab-europe.com\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/dab_logo-main-color-scaled.png\",\"width\":2560,\"height\":1463,\"caption\":\"dab\"},\"image\":{\"@id\":\"https:\\\/\\\/www.dab-europe.com\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2 - dab","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/","og_locale":"en_US","og_type":"article","og_title":"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2 - dab","og_description":"Part 2 \u2013 \u201eCluster\u201c in \u201eACL\u2122 Robotics\u201c This is the second part of a short series of blog posts where we present you two machine learning procedures with the analytic&nbsp;Software ACL Robotics.&nbsp;The \u201cACL\u2122 Robotics\u201d software solution, which has been established on the market for many years, supports the manual and automated analysis of large amounts [&hellip;]","og_url":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/","og_site_name":"dab","article_modified_time":"2024-07-30T13:36:30+00:00","og_image":[{"width":1024,"height":683,"url":"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2024\/07\/G19A0698-1024x683.jpg","type":"image\/jpeg"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"11 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/","url":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/","name":"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2 - dab","isPartOf":{"@id":"https:\/\/www.dab-europe.com\/en\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/#primaryimage"},"image":{"@id":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/#primaryimage"},"thumbnailUrl":"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2024\/07\/G19A0698-scaled.jpg","datePublished":"2020-04-23T00:15:00+00:00","dateModified":"2024-07-30T13:36:30+00:00","breadcrumb":{"@id":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/#primaryimage","url":"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2024\/07\/G19A0698-scaled.jpg","contentUrl":"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2024\/07\/G19A0698-scaled.jpg","width":2560,"height":1706},{"@type":"BreadcrumbList","@id":"https:\/\/www.dab-europe.com\/en\/articles\/the-machine-learning-commands-train-predict-and-cluster-explained-on-a-practical-example-part-2\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.dab-europe.com\/en\/"},{"@type":"ListItem","position":2,"name":"Articles","item":"https:\/\/www.dab-europe.com\/en\/articles\/"},{"@type":"ListItem","position":3,"name":"The Machine Learning commands \u201cTrain\u201d, \u201cPredict\u201d and\u201cCluster\u201d explained on a practical examle part 2"}]},{"@type":"WebSite","@id":"https:\/\/www.dab-europe.com\/en\/#website","url":"https:\/\/www.dab-europe.com\/en\/","name":"dab","description":"","publisher":{"@id":"https:\/\/www.dab-europe.com\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.dab-europe.com\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.dab-europe.com\/en\/#organization","name":"dab","url":"https:\/\/www.dab-europe.com\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.dab-europe.com\/en\/#\/schema\/logo\/image\/","url":"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2025\/02\/dab_logo-main-color-scaled.png","contentUrl":"https:\/\/www.dab-europe.com\/wp-content\/uploads\/2025\/02\/dab_logo-main-color-scaled.png","width":2560,"height":1463,"caption":"dab"},"image":{"@id":"https:\/\/www.dab-europe.com\/en\/#\/schema\/logo\/image\/"}}]}},"_links":{"self":[{"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/articles\/5626","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/articles"}],"about":[{"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/types\/articles"}],"author":[{"embeddable":true,"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/users\/17"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/media\/5585"}],"wp:attachment":[{"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/media?parent=5626"}],"wp:term":[{"taxonomy":"articles_category","embeddable":true,"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/articles_category?post=5626"},{"taxonomy":"articles_tag","embeddable":true,"href":"https:\/\/www.dab-europe.com\/en\/wp-json\/wp\/v2\/articles_tag?post=5626"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}