<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[From Data Analytics to Artificial Intelligence - Medium]]></title>
        <description><![CDATA[A publication to bridge the gap between two buzzwords — Data Science and Artificial Intelligence - Medium]]></description>
        <link>https://medium.com/from-data-analytics-to-artificial-intelligence?source=rss----890b40273153---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>From Data Analytics to Artificial Intelligence - Medium</title>
            <link>https://medium.com/from-data-analytics-to-artificial-intelligence?source=rss----890b40273153---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 03:21:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/from-data-analytics-to-artificial-intelligence" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Seamless Apache Spark Set up for Big Data Analytics]]></title>
            <link>https://medium.com/from-data-analytics-to-artificial-intelligence/seamless-apache-spark-set-up-for-big-data-analytics-210331e1b24?source=rss----890b40273153---4</link>
            <guid isPermaLink="false">https://medium.com/p/210331e1b24</guid>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[scala]]></category>
            <category><![CDATA[apache-spark]]></category>
            <category><![CDATA[jupyter-notebook]]></category>
            <category><![CDATA[apache-zeppelin]]></category>
            <dc:creator><![CDATA[Babatunde]]></dc:creator>
            <pubDate>Mon, 21 May 2018 09:00:30 GMT</pubDate>
            <atom:updated>2018-05-21T17:16:00.678Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>About Apache Spark</strong></p><p>Apache spark is a unified analytics engine for large-scale data processing. In lay man terms, Apache spark is unified in the sense that it allows you to do Batch Processing, Streaming Analytics i.e Real-time analytics, Graph Analytics and Machine Learning all in one engine.</p><p>Also, it handles large-scale data easily because, it’s first a cluster computing framework build to bridge the limitation of MapReduce, and second it expresses a paradigm that makes it fault tolerant, fast and with great in-memory processing.</p><p>Building on top of Map Reduce paradigm (which is the basics of scalable data analysis anyways), spark has come to be a force in the big data world, Allowing for fault tolerant batch analysis with programming languages of choice for data scientist. i.e Python, R and Scala, and the ability to write SQL query on in-memory data. Also, you can also leverage Hadoop’s Hive to query your data in spark using Hiveql a perfect fit for many semi-structured data and that’s a tip of the iceberg on how much you can use spark for your analytics project.</p><p><strong>Challenge of Apache Spark Set up</strong></p><p>Apache spark framework has different components that come together for it’s capability. They include:</p><ul><li>A cluster for distributed processing</li><li>A cluster manager</li><li>Spark core</li><li>A distributed storage system</li></ul><p>Bringing this together to actually provision a real-world engine to try out Apache Spark can be daunting. Thanks to google, you can however provision a real life engine like you will in production on cluster.</p><p>I am going to show how to set up Spark on a 2 or more cluster machine seamlessly using google cloud.</p><p><strong>Google’s Solution</strong></p><p>Let’s get started with what Google has to offer for scalable big data analytics. Their solution is <strong>Google Cloud Dataproc. </strong>According to Google, Cloud Dataproc is a fast, easy-to-use, fully managed cloud service for running <strong>Apache Spark </strong>and <strong>Apache Hadoop </strong><em>clusters</em> in a <em>simple</em>, <em>most cost-efficient </em>way. True to their talk, Dataproc is truly simple to deploy and excellently cost-efficient.</p><p>Dataproc has this to offer seamlessly:</p><ol><li>Apache Hadoop and it’s file system HDFS which uses Google File System backend (Google version of HDFS)</li><li>Apache Spark</li><li>Apache Zeppelin</li><li>Jupyter Notebook and added python packages</li><li>Apache Hive</li><li>Intel BigDL to do Scalable Deep Learning</li><li>Apache Zookeeper</li><li>Apache Kafka (My favourite for Streaming Analytics)</li><li>And some other Hadoop Frameworks, can check the list <a href="https://github.com/GoogleCloudPlatform/dataproc-initialization-actions">here</a></li></ol><p>Let’s dive into the setup.</p><p><strong>Basic Requirement</strong></p><p>A system and Internet connection</p><ol><li>Create a google cloud account <a href="https://cloud.google.com/">here</a> — note you must have a gmail to create a free google cloud account. With this, you are generously given $300 dollar to spend wisely on google’s cloud infrastructures for a year. You can go through the first part of <a href="http://cs231n.github.io/gce-tutorial/">this setup tutorial</a> to create an account and set up a project for your account. <br><strong><em>PS: The setup ends before the topic “Create an image from our provided disk”.</em></strong></li><li>Enable Billing for your project, this you can do by clicking on <strong>Upgrade </strong>after your account has been create, then from the left side menu select <strong>Billing, </strong>then select <strong>Go to linked billing account</strong> to initiate billing for the current project. More details can be found <a href="https://cloud.google.com/billing/docs/how-to/modify-project#enable_billing_for_a_new_project">here</a></li><li>Install google gcloud client. You can follow <a href="https://cloud.google.com/sdk/docs/">this google set-up</a> for installing gcloud client depending on your operating system version.<br><em>PS: I love Ubuntu, Windows now have an Ubuntu interface for windows 10 that you can set up with these tutorials </em><a href="https://tutorials.ubuntu.com/tutorial/tutorial-ubuntu-on-windows#1"><em>here</em></a><em> ,</em><a href="https://docs.microsoft.com/en-us/windows/wsl/install-win10"><em>here</em></a><em> or </em><a href="https://seanthegeek.net/234/graphical-linux-applications-bash-ubuntu-windows/"><em>here</em></a></li><li>Provision you hadoop. Here is the code from the command line that seamlessly provision your hadoop system</li></ol><p>a. Create a cluster name</p><pre>$ CLUSTER_NAME=your_cluster_name</pre><p>b. Create your cluster with some <a href="https://cloud.google.com/dataproc/docs/concepts/configuring-clusters/init-actions">Initialization Actions</a></p><pre>gcloud dataproc clusters create $CLUSTER_NAME \<br>--initialization-actions=gs://dataproc-initialization-actions/bigdl/bigdl.sh,gs://dataproc-initialization-actions/zeppelin/zeppelin.sh,gs://dataproc-initialization-actions/conda/bootstrap-conda.sh,gs://dataproc-initialization-actions/conda/install-conda-env.sh,gs://dataproc-initialization-actions/jupyter/jupyter.sh \<br>--metadata &#39;JUPYTER_CONDA_PACKAGES=&quot;dask numpy pandas matplotlib&quot;,PIP_PACKAGES=pandas-gbq&#39; \<br>--initialization-action-timeout=10m</pre><p><strong><em>PS: </em></strong><em>Initialization actions are the added setups for your cluster and once added, they are automatically set-up with your cluster once it’s created. Initialization actions set up a base for </em><strong><em>Metadata </em></strong><em>which can be thought of as other configurations for the initialization actions. </em><strong><em>Optional Initialization action timeout </em></strong><em>is a time period for the cluster creation to terminate if the initializations haven’t be initialized, I think Google added this to control resource initialization, default if you don’t pass an argument is </em><strong><em>10m</em></strong></p><p>5. Start Gcloud</p><pre>gcloud compute ssh --ssh-flag=&quot;-nND 1080&quot; $CLUSTER_NAME-m</pre><p>And with this your cluster should be created.</p><p>Connect to jupyter notebook and apache zeppelin</p><ol><li>To connect to Jupyter notebook or Apache zeppelin from you locals, it’s simple, the idea is basically tunneling proxy server to your local. The only requirement is that you have chrome or any other browser you will be using installed on your system. <br><em>I prefer Chrome.</em></li></ol><pre>$ chrome.exe --proxy-server=&quot;socks5://localhost:1080&quot; \ <br>--host-resolver-rules=&quot;MAP * 0.0.0.0 , EXCLUDE localhost&quot; \<br>--user-data-dir=/tmp/</pre><p><strong><em>PS: The chrome.exe </em>is the installation path of your Chrome browser e.g for windows will be something like</strong></p><pre>&quot;C:\Program Files (x86)\Google\Chrome\Application\chrome.exe&quot; <br>--proxy-server=&quot;socks5://localhost:1080&quot; \ <br>--host-resolver-rules=&quot;MAP * 0.0.0.0 , EXCLUDE localhost&quot; \<br>--user-data-dir=/tmp/</pre><pre>Optionally - You can customize your web browser to show that it&#39;s using your cluster hence to route your jupyter with cluster name just add this to the above code:</pre><pre>&quot;<a href="http://bigdl-on-zeppelin-m:8123/">http://{insert_cluster_name}-m:8123</a>&quot; \</pre><p>On your cluster server, Jupyter notebook runs on port 8123 by default and Apache zeppelin runs on port 8080, you can now go to your browser and connect via ‘localhost:8080’ and localhost:8123 / {cluster_name}-m:8123.</p><p><strong><em>Congratulations, you just provisioned a cluster to try out big data projects.</em></strong></p><p>In the next article, I will be working through a simple analysis of a Json file in Apache Zeppelin.</p><p><strong>Some Other helpful tips</strong></p><p><strong><em>Ssh into master with</em></strong></p><pre>gcloud compute ssh $cluster_name-m</pre><p><strong><em>Restart any service running in your cluster with systemctl </em></strong>— <strong><em>PS: This is only present in Ubuntu</em></strong></p><p>To restart any of the services, you ssh into the master, list out the services running on your server and restart any of the services with the following commands respectively</p><pre>systemctl list-unit-files --type=service</pre><pre>systemctl restart {service}</pre><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=210331e1b24" width="1" height="1" alt=""><hr><p><a href="https://medium.com/from-data-analytics-to-artificial-intelligence/seamless-apache-spark-set-up-for-big-data-analytics-210331e1b24">Seamless Apache Spark Set up for Big Data Analytics</a> was originally published in <a href="https://medium.com/from-data-analytics-to-artificial-intelligence">From Data Analytics to Artificial Intelligence</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Power BI — The Beginning of Data for many things Intelligence II]]></title>
            <link>https://medium.com/from-data-analytics-to-artificial-intelligence/introduction-to-power-bi-the-beginning-of-data-for-many-things-intelligence-ii-f5550bd6b2a3?source=rss----890b40273153---4</link>
            <guid isPermaLink="false">https://medium.com/p/f5550bd6b2a3</guid>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[sql]]></category>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Babatunde]]></dc:creator>
            <pubDate>Wed, 31 Jan 2018 15:37:43 GMT</pubDate>
            <atom:updated>2018-01-31T15:37:43.304Z</atom:updated>
            <content:encoded><![CDATA[<p>So, we are going to continue the journey of expressing data as the vital thing when making decisions. In the world of Business Intelligence, it is very important to understand your data — this will help you make informed decisions regarding the right approach to structure your dashboard so that decision makers will get the most out of it.</p><p>Like every tool, Power BI comes with it’s features to work with data. It has some cool features that allows you to do great data modelling before you create visualizations. Behind the scene Power BI makes use of <em>Power Query </em>and <em>Power Pivot. </em>These tools were introduced into Excel when Microsoft made the move to be relevant in the Big Data world, since one of their major data product, Excel, have a wide adoption for data manipulation and transformation, these tools help expand Excel to handle querying much larger data and data from different sources. <strong>Power Query </strong>gives Excel the capacity to transform millions more rows of data easily while <strong>Power Pivot </strong>allows Excel to get large data from multiple sources and work with it conveniently. These two tools have also been embedded into Power BI as it’s back-end data collection and data transformation machine.</p><p><strong>1. Sample way of Connecting to a Database with Power BI</strong></p><p>So, in practice, you will need to connect to a Database and get the relevant tables that will be required to build a dashboard. This I will do with a sample Access Database. You can get the data <a href="https://github.com/MicrosoftLearning/Analyzing-Visualizing-Data-PowerBI/blob/master/Lab1/PowerBI%20AccessDB.zip">here</a>.</p><p>Connecting to the database is pretty simple, Under <strong>Get Data </strong>pick <strong>Access Database</strong> and select the folder containing your DB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/582/1*VnhVGsCvRmiTnpa_aI-fOw.png" /><figcaption>Connect to Access Databse</figcaption></figure><p><em>Note: This may require that you have </em><strong><em>Access Database Engine Connector</em></strong><em>, you can download it </em><a href="https://www.microsoft.com/en-us/download/confirmation.aspx?id=13255"><em>here</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/482/1*LUPavunDpqYw9xE4X7LdEw.png" /><figcaption>Connect to the folder containing your database</figcaption></figure><p>After connecting to the database, you will be able to see the tables that are present in the Database.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/422/1*c_XcAB-dkaC-gDmfB7_fCg.png" /><figcaption>Figure contains the list of tables in your Database. You can select each to view before you either <strong>Load </strong>or <strong>Edit.</strong></figcaption></figure><p>You can now go ahead and select the tables you want to work with and either <strong>load </strong>directly into Power BI or <strong>edit </strong>the tables with Power Query, my approach usually is to edit the tables where I can then do some data transformations before sending to Power BI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/623/1*1TeiA2wQ-EYZMJy7c48RqQ.png" /><figcaption>The tables in the database as represented with Power Query</figcaption></figure><p>This is a typical scenario of how you will interact with Databases in practice. Some other points to note:</p><p>a. Once you finish working on your data and you click <strong>close and apply</strong> Power BI automatically detects <em>relationship </em>between your tables, this is very great for visualization. More on Relationship later in next tutorial.</p><p>b. At the right end of the power query pane above is <strong>Applied Steps, </strong>this allows you to go back on some edit you do on your data and you want to take back.</p><p><strong>2. Some Simple Data Manipulation:</strong></p><p>This section makes use of the data used in Tutorial I, this is because it allows us to do some very basic things that form the big idea behind using Power BI for Business Intelligence and Visualizations with a very basic data.</p><p>We will load the csv containing <strong>Mexico sales</strong> only.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/1*5OEM68V8TopWmqF7giMeTg.png" /><figcaption>This is what you should have after loading Mexico sales data</figcaption></figure><p>We can extract out the date, month and year from the date column into separate Columns — Our approach will be :</p><ul><li>Click on the Date Column to ensure it is selected</li><li>Right click on it to make a copy by selecting <strong>Duplicate Column, </strong><em>this is important so that splitting the column will not affect our initial column.</em></li><li>Under <strong>Transform </strong>, Click on the drop-down in <strong>Split Column , </strong>then split column <strong>By delimiter.</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/623/1*PxO8x5AolBYvNbtHd_2aEg.png" /><figcaption>Copy of Data Column and Split Icon with it’s drop-down</figcaption></figure><p><em>Note : — Ensure you observe what you can split your column with, in our case it’s a character ‘/’ that separates our data into Month, Date and Year. Also, the basic delimiters may not contain your separator, after you have checked your column well to know what separates it, you can create a custom delimiter.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/462/1*vegkSlaSYtPFdqgJVxO7ZA.png" /><figcaption>Split column by a <em>custom delimiter ‘/’</em></figcaption></figure><p><em>SN: You can choose where the values in your column splits at, splitting left-most delimiter will only split at the very first delimiter it encounters from the left and same goes for the right-most delimiter button. However in our case, we are splitting at every occurrence of the delimiter.</em></p><ul><li>Click <strong>OK, </strong>and your data will be separated into 3 Columns</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/1*9pBU98yGYhPmyg8SnUsCLw.png" /></figure><ul><li><strong>Rename </strong>the columns — right click each column and select <strong>Rename</strong>, in our case we rename the columns to <strong>Month, Day </strong>and <strong>Year.</strong></li><li>Load your transformed Data with <strong>Close and Apply</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/623/1*KrC9QLD7r0H9hF7Xx95VjQ.png" /><figcaption>This is the view of your data from the ‘Data’ icon in Power BI</figcaption></figure><p>Another transformation we can do is to categorize our months with real month name</p><p>Approach:</p><ul><li>From <strong>Home </strong>bar, select <strong>Edit Query, </strong>this takes you back to the Power Query Pane.</li><li>Select <strong>Add Column, </strong>and then <strong>Conditional Column</strong>. Conditional columns allows you to add a new column based on the condition from another column i.e. say I have a month column denoted with numbers 1–12, I could create another column based on that column by replacing each month number with the actual Month name, January, February…December.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/585/1*ihFZYJWejtoA5BJsX4bnmw.png" /><figcaption>Selecting Conditional Columns</figcaption></figure><ul><li>Conditional Column allows you to add <em>if-else </em>condition rules.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/1*YD-XOHIkjQOVxHmYf-lM4w.png" /></figure><ul><li><strong>Rename </strong>the column and make Visualizations</li></ul><p>A typical visualization on the newly created column will look like this;</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/1*PWHFObfnV-aGseWxS6rtsA.png" /><figcaption>Typical visualization from the newly created condition column</figcaption></figure><p>Note that the month was sorted alphabetically rather than by the actual month sort, we can change this by sorting <strong>Month Name </strong>based on another column and in this case <strong>Month </strong>which contains the <em>number representation</em> for each month.</p><p>Approach:</p><ul><li>Select <strong>Data </strong>icon in Power BI</li><li>Select the column you want to sort from the <strong>Field </strong>in our case <strong>Month Name.</strong></li><li>Select <strong>Modelling </strong>tab and select <strong>Sort by Column</strong> then pick <strong>Month</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/623/1*kJHT5EwzusLGw-Yij8TdSA.png" /></figure><ul><li>Return and check Visualization</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/623/1*vwlkF_wKjgTRTeSdgg5jXQ.png" /></figure><p>A final thing we will like to do with the <em>Mexico sales </em>data is to know how to create new table from an existing table in Power BI.</p><p>Approach:</p><ul><li>Select <strong>New Table </strong>from <strong>Modelling </strong>tab</li><li>In the small space where we have “<em>Table =”, </em>you will create an expression of what you want the new table to have as it’s data. The space is not only used in creating table, it’s used for many calculation in Power BI. The language used for creating this calculation is called <strong>DAX — Data Expression Language. </strong><em>It is a very powerful language that allows you to do Data Transformation on the fly.</em></li><li>You can change the table name to your choice before you create the expression by simply erasing “Table” and adding your own name. In our case we will create a table with the unique Months and their total Revenue.<br>DAX code SUMMARIZE(‘MX Sales’, ‘MX Sales’[Month Name], “Total Revenue”, SUM(‘MX Sales’[Revenue])).<br>What the code does back-end is to create a summary of your initial table, based on a grouping (by Month Name) and a calculation ( a Sum) over your groups. Finally it renames the aggregated column as “Total Revenue”.</li></ul><p>This is a very basic approach to transforming data with Power BI, One vital thing that is often useful again is creating <strong>Measures. </strong>This and many more we will look at in coming tutorials.</p><p>You can go ahead and create some transformations using the <strong>Access DB, </strong>make visualizations and let’s see some of them in the comment section.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f5550bd6b2a3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/from-data-analytics-to-artificial-intelligence/introduction-to-power-bi-the-beginning-of-data-for-many-things-intelligence-ii-f5550bd6b2a3">Introduction to Power BI — The Beginning of Data for many things Intelligence II</a> was originally published in <a href="https://medium.com/from-data-analytics-to-artificial-intelligence">From Data Analytics to Artificial Intelligence</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Power BI — The Beginning of Data for many things Intelligence I.]]></title>
            <link>https://medium.com/from-data-analytics-to-artificial-intelligence/introduction-to-power-bi-the-beginning-of-data-for-many-things-intelligence-i-c6003d839588?source=rss----890b40273153---4</link>
            <guid isPermaLink="false">https://medium.com/p/c6003d839588</guid>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[business-intelligence]]></category>
            <dc:creator><![CDATA[Babatunde]]></dc:creator>
            <pubDate>Tue, 23 Jan 2018 02:24:05 GMT</pubDate>
            <atom:updated>2018-01-23T02:45:23.374Z</atom:updated>
            <content:encoded><![CDATA[<p>Power BI is one of the many tools that is used in Business Intelligence. Remember what Business Intelligence is? It is an approach wherein you use software tools to make compelling insight out of data for Decision makers.</p><p>So what Power BI helps you do in this case is simple, consider your organization, a data-driven one, or let’s say your organization is just getting to use data to solve it’s many problems, one key thing that company use to drive growth is to monitor KPI — Key Performance Indicator. Remember what KPI is? These are the unique metrics that organizations track to monitor growth, it could be ratio of retention to customer acquisition, could be revenue per employee and many other metrics. But one thing stands out for KPI, this is that they are the life metrics organizations monitor to breath. You can refer back to the slide for details of things around Business Intelligence and KPI.</p><p>Power BI is one of the tools available for doing Business Intelligence and the end product of works with Power BI is to create a Business Intelligence Dashboard which managers and decision makers can look through in seconds and make informed Decision.</p><p>Where Data and Data Scientist come into play for this is;</p><p>· The Acquisition of Data</p><p>· Manipulation of Data</p><p>· Modelling the Data</p><p>· Creating compelling Visualizations from the Data</p><p>· Guiding Decision makers to understand the visualizations.</p><p>Power BI allows a Data Scientist to do this on the fly in a somewhat behind the scene process and make Decision makers happy. This short tutorial is to get you into the feel of how Power BI does this.</p><p>What you have to look out for is the process by which you use the tool and it’s features to achieve your dashboard. We will be looking basically at the two points i.e Acquisition and simple Visualisations of Data in this tutorial and in further tutorials we will conclude the other points.</p><ol><li>Data Acquisition</li></ol><p>In Real life, Data is usually gotten from a Database or downloaded as a file and because you might need to join multiple Data together, let’s just say we see our data coming from a Database with multiple Table. For our example we will first get a csv file into power BI</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/485/1*upkE8WdJN_TPgOdnoBkUXA.png" /><figcaption>The Get Data features allows you to get Data from a lot of sources and data types into power BI</figcaption></figure><p>The circled portion is the feature that let’s you get data from many sources into Power BI. If you click the drop down and then <strong>More, </strong>you see a list of data sources that can be Power BI can read data from. You could scroll down to see more on the application.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*bVsHKE0r9LwhOjjrshN3Tw.png" /><figcaption>Once you pick the source of your data then click on connect</figcaption></figure><p>In our case we are getting a CSV file and then we hit <strong>Connect</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/853/1*lZI1sDSEZWTjlxaYuEo4PA.png" /><figcaption>Takes you to the source of your data and you can access whichever table you want, in our case the International Folder contains our “tables”.</figcaption></figure><p>And after you connect with the data you want to load it in, then you see a view such as this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/870/1*zBYb7jqHgdYoHiKxhs5JAw.png" /><figcaption>Panel to view your table before you load it into your workspace</figcaption></figure><p>This panel is for you to view your data, understands it and relate with it before you load it into your work-space for visualisation.</p><p>Then you proceed to loading your data into the workspace. Some things come alive on your dashboard once you load the datasets which are.</p><p>a.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/259/1*ad4HUE-si_zb72Etp34-0A.png" /><figcaption>Fields that shows the Table — labelled as A and the columns in the table labelled as B.</figcaption></figure><p>b.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/707/1*7MRt7CJCWnB8_Exbrf3bRw.png" /><figcaption>.The circled field below that is on your Dashboard which allows you to view the Tables that have in your Fields.</figcaption></figure><p>This concludes loading Data into Power BI.</p><p>2. A very basic Visualization</p><p>This Simple virtualisation forms the root of all dashboards you will creating with Power BI.</p><p>Couple of things to note: Consider a typical work-space termed <strong>Report </strong>in Power BI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aKZwLo8ImowPVuMkEzMRwg.png" /></figure><p>a. Reports Icon: This icon takes you to the ‘Workspace’ where you build the dashboards i.e Visualisations of the metrics and things you want to see from your data</p><p>b. Visualisations: Comes bundled with the Reports Icon, you are able to pick whichever visualisation type you want to make use of, this usually depends on <em>the kind of questions you are willing to answer from your data</em></p><p>c. A list of check box next to the columns contained in the data: This allows you to check which columns you are making your visualisations with. Again this depends on <em>the kind of questions you are willing to answer from your data </em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JjjUCRM6lug4BKzXPULG9Q.png" /><figcaption>Revenue by Zip Code. Take careful note of the ticked boxes in the fields.</figcaption></figure><p>This marks the creation of your first minimal dashboard.</p><p><strong>Task: </strong>Let’s say I want to observe revenue for my products over two countries.</p><ol><li>Load the multiple files in your folder through Power BI load by Folder</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/572/1*SdotoNwRtkxJ1RgXCz6y4g.png" /><figcaption>Use the Folder source to get multiple tables in a folder</figcaption></figure><p>2. Preview, then click <strong>combine and Edit</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/917/1*dlFtc_ub4fUjQDPG-OFiuw.png" /><figcaption>Combine and Edit can be seen when you click the drop down in the combine icon</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/845/1*3BIUZ0ZCN4O5Ijceo3Dgkw.png" /><figcaption>Shows the preview</figcaption></figure><p>3. On the edit pane shown below — and because we want to only make our visualisation for 2 countries, Mexico and Germany, we will filter off the remaining countries</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NsuDel09tjJCgHSJ8oVpZg.png" /><figcaption>The Drop down circled allow you to be able to see distinct values of the column and then be able to filter it.</figcaption></figure><p>4. Once you click on the drop down, you will be given the option to <em>load more </em>if it’s only one of the distinct values you are seeing. And after loading more you are able to un-select the values you don’t want to see</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*pduTLfmu9phBznOjmENhDQ.png" /></figure><p>5. After which you click OK, then <em>Close and Apply </em>at the left upper part of the image above — with this you only have data for Germany and Mexico in your Report/Workspace.</p><p>6. Visualising the task</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xacxFbPa7Q5WPd7H6gQ3tw.png" /><figcaption>This shows the revenue by product ID for each country and the sum total for the two countries</figcaption></figure><p>The above shows first, the total revenue for both country for each product ID, then shows a differentiation of what each country contributes to the total revenue observed. The above is an example of a <strong><em>Stacked Histogram </em></strong>. Couple of things to note</p><p>a. The kind of visualisation picked</p><p>b. The fields</p><p>c. Axis, Legend and Values: Axis is usually your x-axis while values is the sum of Revenue in our case. Legend is more of a differentiating value hence Country in our case.</p><p><strong>Task</strong>: Replicate this for 3 and 4 Countries in the data.</p><p>PS: Data can be gotten from <a href="https://github.com/MicrosoftLearning/Analyzing-Visualizing-Data-PowerBI/blob/master/Lab1">International Sales File from this GitHub repository</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c6003d839588" width="1" height="1" alt=""><hr><p><a href="https://medium.com/from-data-analytics-to-artificial-intelligence/introduction-to-power-bi-the-beginning-of-data-for-many-things-intelligence-i-c6003d839588">Introduction to Power BI — The Beginning of Data for many things Intelligence I.</a> was originally published in <a href="https://medium.com/from-data-analytics-to-artificial-intelligence">From Data Analytics to Artificial Intelligence</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Data to AI: How Data is is the fuel moving the Artificial Intelligence Wave.]]></title>
            <link>https://medium.com/from-data-analytics-to-artificial-intelligence/from-data-to-ai-how-data-is-is-the-fuel-moving-the-artificial-intelligence-wave-c41148fd893b?source=rss----890b40273153---4</link>
            <guid isPermaLink="false">https://medium.com/p/c41148fd893b</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[machine-intelligence]]></category>
            <dc:creator><![CDATA[Babatunde]]></dc:creator>
            <pubDate>Mon, 15 Jan 2018 10:27:29 GMT</pubDate>
            <atom:updated>2018-01-31T15:59:14.728Z</atom:updated>
            <content:encoded><![CDATA[<p>This article is intended to explain a little on a matter that is pertinent to the Artificial Intelligence community, this is that from Data Science to Artificial Intelligence, Data is the fuel. Taking a quick delve into this matter, I will pick briefly from the knowledge and words of Andrew Ng an awesome AI role model for many, he has explicitly said in many of his talk that Data is the new oil. This is so because the recent advances in Deep Neural Network models which are the backbone of most Artificial Intelligence breakthrough stem from the realization of the fact that with more data these models work incredibly well.</p><p>Dissecting further into what Ng means, right now businesses are massively using Machine Learning to make business decisions in predicting fraud models, risk management and smart business intelligence to mention but a few. They are harnessing a lot of customer data to generate a better customer experience with the end result of making more money and it’s working.</p><p>Taking a closer look at the concept of Artificial Intelligence relative to data!</p><blockquote>The term “artificial intelligence” is applied when a machine mimics “cognitive” functions that humans associate with other human minds, such as “learning” and “problem solving” — <strong>Wikipedia</strong></blockquote><p>This, in simple terms means, we try to make machines behave like humans. Looking critically at how humans learn. We perceive our environment, make mistakes from it, learn from those mistakes and mistakes of others and we become wise along the way. The curious reader can check out this wonderful <a href="https://sites.google.com/site/humancenteredactionresearch/how-humans-we-learn">article</a>.</p><p>Some few points I am extracting from the article:</p><ul><li>Our capacity to learn is largely determined by the level of our conscious attention to our senses. The more we are aware of what we see, smell, taste, hear, touch, or gather from the tone or mood of a situation, the more likely we will process information on a thoughtful level.</li><li>central nervous system rapidly gathers, organizes, interprets, and makes sense of the inputs, to prepare our body and mind (peripheral nervous system — sensory neurons, clusters of neurons, and nerves) to adapt and take action based on need or circumstance.</li><li>Our working memory “draws” from long-term memories to seek connections, make new meanings, create mental visualizations, and recognize familiar patterns, which in turn prepares the brain to establish relationships, organize information, create categorize, and consider new understandings.</li><li>When learning environments and conditions engage multiple connections to the brain, humans are more likely to make greater attempts to process, take action, and apply new learnings.</li></ul><p>I will take turns to explain this four points in the concept of Artificial Intelligence and it’s reliance on Data.</p><ol><li>Machine learning algorithm, when exposed to the right set of data and features, has proven time without number to produce incredible results. At the other end, machine learning when fed with untreated data returns — well garbage in, garbage out.</li><li>Machine Learning models which are currently making waves in the world of Artificial intelligence have a pre-processing pathway that tends to organize the data before the Algorithms learn. Okay, cutting the shit, Machine learning algorithms are a bunch of vector applications that pick up salient part of a data and come up with recognition around the pattern that describes the dataset. Hence, they are supersmart in pattern recognition when the data is well organized.</li><li>Lately, badass Algorithms that are mainstream in core Artificial Intelligence — talking about Deep Learning Algorithms have learnt to retrain stuffs for particular amount of time, transitioning that memory to realizing many new things to come. It’s supercool.</li><li>And on the final note, when algorithms have finally learned from this data, they do well on new data or let’s put it in the context of point 4, they do well in new environments!</li></ol><p>In all these, we have seen that data is the bedrock of Machine Learning through to Deep Learning that lines latest development in Artificial Intelligence. In further articles, we will be dissecting principles of Machine Learning and Artificial Intelligence through the fundamental concept of “Learning from Massive Data”</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c41148fd893b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/from-data-analytics-to-artificial-intelligence/from-data-to-ai-how-data-is-is-the-fuel-moving-the-artificial-intelligence-wave-c41148fd893b">From Data to AI: How Data is is the fuel moving the Artificial Intelligence Wave.</a> was originally published in <a href="https://medium.com/from-data-analytics-to-artificial-intelligence">From Data Analytics to Artificial Intelligence</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>