<?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[Stories by Dataleon on Medium]]></title>
        <description><![CDATA[Stories by Dataleon on Medium]]></description>
        <link>https://medium.com/@dataleonlabs?source=rss-7c9d738e491f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*5wda_31McgryblT2lCbsyQ.png</url>
            <title>Stories by Dataleon on Medium</title>
            <link>https://medium.com/@dataleonlabs?source=rss-7c9d738e491f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 07:45:10 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dataleonlabs/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[What Is The Difference Between Image Classification & Object Detection Techniques?]]></title>
            <link>https://medium.com/@dataleonlabs/what-is-the-difference-between-image-classification-object-detection-techniques-7f23036d18b7?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/7f23036d18b7</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[classification-algorithms]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[computer-vision]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Tue, 16 Nov 2021 18:31:51 GMT</pubDate>
            <atom:updated>2021-11-16T18:31:51.358Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-uEkF0e1grvv8N8RMzpCvg.jpeg" /></figure><p><em>Image classsification and object detection techniques are important methods used in computer vision. These techniques help machines understand and identify real-time objects and environments with the help of digital images as inputs. Since these two methods turn around identifying objects in digital images, it often leaves confusion: what image classsification and object detection techniques actually are and how they differ from each other?</em></p><p>In this article, we will learn about these two techniques and discuss what makes them different.</p><h4>Image classification</h4><p>In simple terms, <a href="https://analyticsindiamag.com/benchmark-analysis-of-popular-image-classification-models/">i</a>mage classification is a technique that is used to classify or predict the class of a specific object in an image. The main goal of this technique is to accurately identify the features in an image.</p><p>Below you may see several types of image classification techniques:</p><ul><li>Image classification based on the information acquired from different sensors;</li><li>Image classification based on the nature of the training sample used in classification;</li><li>Image classification based on the basis of the various parameter used on data;</li><li>Image classification based on the nature of pixel information used on data;</li><li>Image classification based upon the number of outputs generated for each spatial data element;</li><li>Image classification based upon the nature of spatial information.</li></ul><p>The extensive amount of time required during the training phase and big data are the main drawbacks of image classification technique.</p><h4>Object detection</h4><p>The aim of object detection is to determine where objects are located in a given image such as object localisation and which category each object belongs to, i.e. object classification. In simple words, object detection is a type of image classification technique, and besides classifying, this technique also identifies the location of the object instances from a large number of predefined categories in natural images. Object detection techniques can be used in real-world projects such as face detection, pedestrian detection, vehicle detection, traffic sign detection, video surveillance, etc.</p><p>Over the past few years, great success has been achieved in a controlled environment for object detection problem. However, in uncontrolled places, in particular, when objects are placed in arbitrary poses in a cluttered and occluded environment, the problem remains unsolved.</p><p><strong>Dataleon</strong> can help you manage your decisions and analyze your<br>images and documents by using powerful Machine Learning algorithms. Its powerful computer vision platform allows you to train both classification and detection models.</p><h3>Dataleon`s contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7f23036d18b7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to train a model to process a contract?]]></title>
            <link>https://medium.com/@dataleonlabs/how-to-train-a-model-to-process-a-contract-6dbffc800548?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/6dbffc800548</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[contracts]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[model-training]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Wed, 20 Oct 2021 07:56:47 GMT</pubDate>
            <atom:updated>2021-10-20T07:56:47.344Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*OeuL0K3e26qO7JCN_HSffQ.png" /></figure><p><em>Contract review is a repetitive and tedious job because junior law firm associates always need to identify the same data points in any given contract: What is effective date of the contract? What are the renewal terms? Who are the parties involved in this contract? Machine learning models can learn to automatically extract and identify key values from contracts, thus saving hundreds of hours of manual labour.</em></p><p><strong>Dataleon </strong>is a no-code AI platform which provides the best ML tools for automation and data processing. It can facilitate the process of training a model by using powerful machine learning algorithms. Before training and testing a model, it is crucial to prepare a significant volume of high-quality data. If the data isn`t clean, doesn`t make sense or has a lot of missing information, it is important to comb through the data, label and prepare it before training the model.</p><h4>Creating a preset.</h4><p>When just starting out with <strong>Dataleon</strong>, you will be asked to create a preset of your data. <em>Presets </em>is the part in which data is collected, prepared before a project creation and finally cleaned out. It may be regarded as a source of data which will be later used for labeling.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wdejAahr8IXMG7D5FOZOKw.png" /></figure><p>After uploading the images within one preset, you will have a possibility to check the images for duplicate or similar files. For this the platform suggests a <em>Scan </em>feature. Once you click <em>Scan</em>, the system will remove duplicate or similar files automatically.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2opPb8KA9HTLAG79NDKfLA.png" /></figure><p>You can also view all the details about every image which you have uploaded. Just click on the image you are interested in and view all its details. You can crop the image, if you are not satisfied with its background, and you can also delete the image.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rWLsZKArbY6ARmyMYMqeQA.png" /></figure><h4>Creating a project.</h4><p>Once you have uploaded and cleaned out your data, you may proceed with a project creation. <em>Projects </em>part is designed for real data labeling, synthetic data generation, and data classification and detection.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sZeC6LfBSAccuftNJP6e1w.png" /></figure><p>If you are certain that the data is accurate, relevant and sufficient, you can use our out-of-the-box Labeling Editor to label your data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nqyz0FU-nC9T5zxUjYRrnA.png" /></figure><p>The large amount of unlabeled data is a growing problem in machine learning. This leaves data scientists with more data than they are capable of analyzing. That’s where active learning comes in. <em>Active learning</em> is the subset of machine learning in which a learning algorithm can query a user interactively to label data with the desired outputs. You can apply <em>Active learning</em> in the <em>Labeling Editor</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JGoiLS9qIc8PDJZ1WqnglQ.png" /></figure><p>The quantity and quality of your training data determine the accuracy of the performance of your machine learning model. Therefore, lack of relevant and accurate data can be another great problem in machine learning. Hopefully, it can be easily resolved thanks to our <em>Scene Editor, </em>i.e. an interface used for data generation<em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*URAsvu6Tny_ven2-1A533w.png" /></figure><h4>Creating and training a model.</h4><p>After you data is collected and cleaned out, you can pass on to the <em>Knowledge Vision </em>section. In this part of the platform you can train your model with your data (either personal or generated in Datalabs) and visualize the result.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sS2sso9Tbngu0su97zGo_A.png" /></figure><p>At <strong>Dataleon </strong>we are making it possible to train a model to process a contract. If you are in need of such a model, just let us know.</p><h3>Dataleon`s contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6dbffc800548" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[New no-code challenge AI]]></title>
            <link>https://medium.com/@dataleonlabs/new-no-code-challenge-ai-792699cd32eb?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/792699cd32eb</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[low-code]]></category>
            <category><![CDATA[no-code]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Mon, 11 Oct 2021 10:27:19 GMT</pubDate>
            <atom:updated>2021-10-11T10:27:19.662Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VQzZEUGyrDGllUvQTvkiGQ.jpeg" /></figure><p><em>The process of building and deploying artificial intelligence (AI) can be long and expensive. It leads organizations to spending an average of around </em><a href="https://appen.com/whitepapers/driving-roi-through-ai"><em>$40 million</em></a><em> a year on AI, while relatively few projects make it into production. Such Automated Machine Learning tools (AutoML) as Microsoft Azure can be a powerful way for technical AI professionals to speed up their workflows. Though, they are expensive, time-consuming, inaccessible to non- technical people and difficult to deploy and maintain.</em></p><p>No-code AI platforms like <strong>Dataleon</strong> are democratizing machine learning by making AI:</p><ul><li><strong>Easy to use</strong>: <strong>Dataleon </strong>allows you<strong> </strong>to start building models straight from datasets without writing code or using complex tools.</li><li><strong>Easy to learn</strong>: <strong>Dataleon </strong>makes it possible to bring artificial intelligence to general public.</li><li><strong>Affordable</strong>: <strong>Dataleon </strong>charges for the number of actions you take per month, so you pay only for what you use.</li><li><strong>Deployable</strong>: <strong>Dataleon</strong>`<strong>s </strong>models can be deployed in any setting in just a few clicks.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Fv4nEPbg2mKvfvVw_SbMZQ.png" /></figure><p>Let’s explore these points in-depth to see why no-code AI platforms can be more efficient than traditional AI tools.</p><h3>No-code AI doesn’t need experts</h3><p>No-code AI tools like <strong>Dataleon </strong>can be used to build and deploy models in minutes, and you’ll be well-grounded in the functionality in hours, as opposed to years.</p><h3>No-code AI platforms are affordable</h3><p>With <strong>Dataleon</strong>, there’s never a model training cost. You can use <strong>Dataleon </strong>API to serve predictions in any setting.</p><h3>No-code AI platforms are fast</h3><p><strong>Dataleon`s</strong> model training process is fast. Thence, it helps deliver the product to the target market 10x faster.</p><h3>No-Code AI is Easy to Deploy and Maintain</h3><p>No-code AI platforms like <strong>Dataleon </strong>allow users to avoid time spent on research. Moreover, no configuration machines are needed for the deployment. As even non-technical professionals can easily deploy and maintain models, it becomes much easier to bring artificial intelligence projects to production.</p><h3>No-code AI is highly adaptive</h3><p>Using tools like <strong>Dataleon</strong>, it’s fast and easy for business users to create ML models. Such a tool can be a great resolution in case you are not sure that your features can work with artificial intelligence.</p><h3>No-code AI platforms are time- and money saving</h3><p>No-code AI tools like <strong>Dataleon </strong>make it possible to avoid machine learning ops systems and development ops systems. Business user can easily avoid deploying models on production.</p><p>At <strong>Dataleon</strong>, we’re focused on making machine learning effortless and affordable for everyone. Our no-code AI platform allows anyone to start using the power of artificial intelligence to automate their business.</p><h3>Dataleon`s contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=792699cd32eb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data preparation on Dataleon platform]]></title>
            <link>https://medium.com/@dataleonlabs/data-preparation-on-dataleon-platform-11df775ecb6e?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/11df775ecb6e</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[labelling]]></category>
            <category><![CDATA[model-training]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Mon, 04 Oct 2021 09:16:22 GMT</pubDate>
            <atom:updated>2021-10-04T09:16:22.123Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J6ZVLll_1ykaarBhQbv4Lw.png" /></figure><p><em>Machine learning is an application of artificial intelligence (AI), that provides system with an ability to automatically learn and improve from experience without being explicitly programmed. ML focuses on the development of computer programs that can access data and use them to learn for themselves. One of the most crucial aspects ML depends on is </em><strong><em>data</em></strong><em>.</em></p><p>On Dataleon we realize that data preparation is a very important step in the machine learning process. Basically, it includes a set of procedures of getting ready the data for training, testing and implementation of an algorithm. This multi-step process involves <strong>data collecting</strong>, <strong>cleaning, validation, transformation </strong>and <strong>labeling</strong>.</p><p>The process of <strong>data preparation</strong> starts with searching for the right data. It involves collecting the data that is believed to be useful in making a prediction and clearly defining the form the prediction will take. It may also embrace talking to project managers and other people with deep expertise in the domain. Deep understanding of our customer`s needs detemine the data we will later use for ML.</p><p>When <strong>gathering the data</strong>, the main problems we face are lack of data, poor quality data and unbalanced data. To solve these problems, Dataleon`s experts use Scenes Editor. It is an interface which is used for data generation. The output data can be later used for labeling.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7az_OgUUuxyV0mvw_GhHqA.png" /></figure><p><strong>Data cleaning</strong> is the next step in data preparation. On this step we remove all the data that do not belong to the dataset. This process involves fixing or removing incorrect, corrupted, incorrectly formated, duplicate, or incomplete data withing a dataset.</p><p><strong>Data validation</strong> takes place after cleaning the data. On this step we check the data for:</p><ul><li>Validity. The degree to which the data conforms to defined<br>business rules or constraints.</li><li>Accuracy. Ensure that your data is close to the true values.</li><li>Completeness. The degree to which all required data is<br>known.</li><li>Consistency. Ensure that your data is consistent within the same<br>dataset and/or across multiple data sets.</li><li>Uniformity. The degree to which the data is specified using<br>the same unit of measure.</li></ul><p>Being on the <strong>data transformation</strong> stage, we convert data from one format or structure into another one. Transformation process can also be referred to as data wrangling, or data mungling, transforming and mapping data from one “raw” data form into another format for warehousing and analyzing.</p><p>The final stage of the data preparation process is <strong>data labeling</strong>. In machine learning it is the process of identifying raw data (images, text files, videos, etc.) and adding one or more meaningful and informative labels to provide context so that a machine learning model can learn from it. Dataleon`s experts use Labeling Editor for data labeling.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CgWGPIvzQMIRsMrOW33IQw.png" /></figure><p>Dataleon API can guide you trough the whole process of data preparation. If you are interested in our services, just let us know.</p><h3>Dataleon`s contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=11df775ecb6e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Training a model on Dataleon platform]]></title>
            <link>https://medium.com/@dataleonlabs/training-a-model-on-dataleon-platform-66d5bf31ae86?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/66d5bf31ae86</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[model-training]]></category>
            <category><![CDATA[labelling]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Fri, 24 Sep 2021 13:30:17 GMT</pubDate>
            <atom:updated>2021-09-24T13:30:17.413Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*BZDMEJ0UnUTeKgPhcwRxNg.png" /></figure><p><em>Training a model simply means </em><strong><em>learning (determining) good values for all the weights and the bias from labeled examples</em></strong><em>. In supervised learning, a machine learning algorithm builds a model by examining many examples and attempting to find a model that minimizes loss; this process is called empirical risk minimization.</em></p><p>On <strong>Dataleon</strong> platform you can train a model in the <strong>Knowledge Vision</strong> part. On the whole the platform consists of four parts. The other three parts of the platform are <strong>Pre-Trained Models</strong>, <strong>Datalabs</strong> and <strong>Analytics</strong>.</p><p>In <strong>Pre-Trained Models</strong> part of the platform you will find our ready-to-use models for data detection, extracton, analysis and classification.In <strong>Datalabs</strong> section you can create data in order to train your own model. In <strong>Knowledge Vision</strong> part of the platform you can train your model with your data (either personal or generated in Datalabs).And finally in <strong>Analytics </strong>section you can analyze your inference request on production.</p><p>In order to train a model you will need to upload some data. These data will be further used for training the model. You can easily do it in <strong>Datalabs/Presets</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*obqbBbDckvkprcTaNi3NRA.png" /></figure><p>Afterwards you can create a project in <strong>Datalabs/Projects</strong> with these data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Kom4Ybiobpxb4iL7W2esCw.png" /></figure><p>When creating a project in <strong>Datalabs, </strong>you can label the data previously uploaded in <strong>Presets </strong>subpart. Below you can see a screenshot of <strong>Labelling Editor</strong>, i.e a labelling interface used to label your datasets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*41SHGBy1yr29ZI0qk5YatQ.png" /></figure><p>In case you do not have any data for labelling , you can generate them in <strong>Scenes Editor</strong>. These data can be also used later for labelling.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e-qDoN5a9ZxBPeFT0IbvOQ.png" /></figure><p>When the project is created you can view all the information about it in <em>Settings</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3AQ46wrN7jCXwPPgAyI6Sg.png" /></figure><p>After working with data, you can finally create and train a model.<strong> </strong>To do this you will be asked to follow 6 steps in <strong>Knowledge Vision</strong> section.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_jYdzErIH6fmqknRnps80Q.png" /></figure><p>On step 4 you will be asked to select either a <em>basic</em> or an <em>expert mode</em>. If your are not a data scientist, select a basic mode. If you select an expert mode, you will need to specify your model and its training parameters, hyperparameters, data augmention and evaluation metrics.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_h5MuOJWiksSsfJuVekn0A.png" /></figure><p>In the process of training you can view all the detailed information, such as logs, checkpoints and metrics.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RHPaMHSGRgxNbH-CqIp2Bg.png" /></figure><p>After that you click <strong>Upload </strong>to upload your data and you can test the model with your real data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4Sf0kNE2g4eIFe1uTSCVPw.png" /></figure><p>Training a model can be easy. Just try!!!</p><p>So, why to choose Dataleon platform for training a model?</p><ul><li>No MLOps or DevOps are needed for deployment.</li><li>You can select <em>Expert mode</em> if you want to.</li><li>Dataleon is no-code AI platform, which designs a custom solution to meet your demands.</li></ul><h3>Dataleon`s contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=66d5bf31ae86" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[5 best platforms for computer vision in 2021]]></title>
            <link>https://medium.com/@dataleonlabs/5-best-platforms-for-computer-vision-in-2021-fa5c51d22a3?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/fa5c51d22a3</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Mon, 20 Sep 2021 08:59:06 GMT</pubDate>
            <atom:updated>2021-09-20T08:59:06.681Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tPWKNUKD-6YLGn21eFAncg.png" /></figure><p><em>According to many research firms such as Gartner, Forrester or MarketsAndMarkets, the global computer vision market is set to reach a total of 20 billion dollars by 2025. Two trends explain this significant growth. On a technological level, deep learning changed the strategy and the computing-related costs have rapidly decreased. On a business level, companies are doubling down on their AI investments after spending time on digital transformation. Computer vision platforms are in demand now, so we have decided to review the best ones. We have created this article so that you could learn more about each platform.</em></p><h3>1. Dataleon</h3><p>Dataleon helps you manage your decisions and analyze your images and documents by using powerful ML algorithms. The platform provides the best ML tools for automation and data processing.</p><p><strong>Platform`s advantages</strong>:</p><ul><li>Dataleon`s experts are at your disposal to design a tailored product and offer you the best customer experience.</li><li>The models are constantly learning to better understand your needs, identify and extract the data that matter to your business.</li><li>Your files are convertible into structured data in less than 2 seconds. Get the results in real time.</li></ul><h3>2. Cogniac</h3><p>Cogniac`s Visual Intelligence Platfrom delivers superhuman performance by virtue of the latest developments in AI and convolutional neural networks.</p><p><strong>Platform`s advantages:</strong></p><ul><li>The drag and drop nature of the platform allows users to focus on the most valuable tasks without developers engagement.</li><li>Cogniac`s platform can derive value from as little as 100 labelled images.</li><li>Cogniac`s AI behaves like a data scientist — optimizes models and course-corrects in real time, greatly reducing setup and implementation, saving hundreds of thousands of dollars.</li></ul><h3>3. Deepomatic</h3><p>Deepomatic is a leading Vision application platform that helps customers get the best out of their field operations.</p><p><strong>Platform`s advantages:</strong></p><ul><li>The no-code and intuitive interface allows business organizations to benefit from computer vision without development skills.</li><li>Your own set of rules and business logic can be integrated in the platform.</li><li>Your applications are always up-to-date thanks to Continuous Improvement of Deepomatic Visual Automation platform.</li></ul><h3>4. Clarifai</h3><p>Clarifai is the leading deep learning AI platform for computer vision, natural language processing, and automatic speech recognition.</p><p><strong>Platform`s advantages:</strong></p><ul><li>The platform integrates labeling, search, training and prediction capabilities.</li><li>It has the largest collection of pre-trained models.</li><li>You can get started with 5000 free operations each month.</li></ul><h3>5. Viso</h3><p>Viso is all-in-one platform to build computer vision applications without code.</p><p><strong>Platform`s advantages:</strong></p><ul><li>Viso lets you create real-time computer vision applications with pre-built modules.</li><li>It allows you to deploy your real-time applications to computers and to process the video feeds of digital cameras in real-time.</li><li>Viso Suite makes it easy to vizualize and monitor your applications in cloud dashboards with interractive charts and dynamic filters.</li></ul><p>Some of the platforms have unique, out-of-the-box tools used for labelling and training. Below you may see a table with the platforms and the tools names.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/675/1*D07qLNxdIWJleoCMbicAoA.png" /></figure><p>We have also compared the platforms by the following criteria:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*ur5UUZ_hT2vqFllAqeNSZg.png" /></figure><p>If your company has an issue and you want to resolve it in the quickest way, just get in touch with us. Dataleon`s team works hard to develop automative innovation and adjustable solutions available in the cloud.</p><h3>Dataleon`s contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fa5c51d22a3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Automated the data extracting from a driver license]]></title>
            <link>https://medium.com/@dataleonlabs/automated-the-data-extracting-from-a-driver-license-b6a5dbffd18f?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/b6a5dbffd18f</guid>
            <category><![CDATA[automation]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ocr-software]]></category>
            <category><![CDATA[drivers-license]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Tue, 01 Sep 2020 10:55:11 GMT</pubDate>
            <atom:updated>2020-09-01T14:03:23.980Z</atom:updated>
            <content:encoded><![CDATA[<h3>Automated data extracting from driver’s license</h3><h4>Case study | driver’s license | OCR | AI</h4><h3>Overview</h3><blockquote>Scanning your driver’s license with OCR</blockquote><h3>Context</h3><p>Your driver&#39;s license is one of the most important documents when you get a car. You need it also to pass control, open a car insurance, or buy a car. Moreover, this document can also be used to prove your identity at the same level as an ID card. In this use case, we will analyze how OCR can help with extracting data from driver licenses.</p><h3><strong>Challenge</strong></h3><ul><li>Extracting information from driver licenses using OCR technology.</li><li>Add possibility to search, edit, and classify the document or the data inside of it.</li></ul><h3>Solution</h3><ul><li>Connect OCR API to the company databases.</li><li>Train the OCR to recognize every type of data from driver licenses.</li><li>Recognize, collect, and classify information.</li><li>Enable converting and storing information in databases.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*98xVnQnuZMEpCJy_uT0VtQ.png" /></figure><h3><strong>Result</strong></h3><ul><li>An automated workflow that helps enterprises to validate identity.</li><li>Saving time on the information reading and extracting processes.</li></ul><p>Young App is a company that aims to guide enterprises in their digitizing journey with OCR. We help companies by improving their productivity with process automation. If you want to know how OCR can help your organization, we recommend you to explore <a href="https://youngapp.co/blog">other use cases</a> or <a href="https://youngapp.co/">to contact us</a>.</p><h3>Young App contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b6a5dbffd18f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From AI to OCR: how does it work?]]></title>
            <link>https://medium.com/@dataleonlabs/from-ai-to-ocr-how-does-it-work-5e1254780caf?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/5e1254780caf</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[artificial-neural-network]]></category>
            <category><![CDATA[ocr-software]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Thu, 27 Aug 2020 08:51:59 GMT</pubDate>
            <atom:updated>2020-08-27T08:51:59.056Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><blockquote>The AI concept took an entirely new role in the modern world. While some managers fear these two words, other executives see in AI the must-have-tool. Indeed, even<strong> Forrester predicts the industry will continue to grow from $250 million in 2016 to $12 billion in 2023</strong>. Beyond these optimistic statistics, companies realised that AI can benefit the productivity of their activity. With the help of machine learning or deep learning, the tools such as<a href="https://www.unite.ai/what-is-robotic-process-automation/"> robotic process automation</a>‎, voice recognition, or <a href="https://medium.com/@youngapp/top-5-ocr-tools-4e299e38ba72">OCR have seen the day</a>. But, how do we go from AI to OCR? And what role does machine learning play in this whole process?</blockquote><figure><img alt="How does artificial intelligence work?" src="https://cdn-images-1.medium.com/max/960/0*g1bVsUpp_xQ4k2Jd.jpg" /></figure><h3><strong>AI and ML, what is the connection?</strong></h3><h4><strong>Artificial intelligence</strong></h4><p>If we explain it as a simple way, we can say that artificial intelligence (AI) is <strong>the science that studies ways to build smart algorithms or machines</strong>. These are created in order to respond to a problematic situation, which usually requires a human brain to be solved.</p><p>However, AI remains much bigger than this short definition. Indeed, AI has developed itself over time, and continues to evolve. A simple chess-playing program, for example, was considered as a form of AI years ago. And, for a good reason, AI has gone far beyond chess-game strategies and overcame many skills that only human beings can develop. And nowadays, AI takes the place of the chess queen in the world of invention.</p><p>That is why tools such as Siri, Alexa, or Google Home came to light. All these new technologies have become strong actors of our daily life and <strong>machine learning rushingly appeared on this dynamic landscape</strong>.</p><h4><strong>Machine learning</strong></h4><p>Often confused with AI, machine learning (ML) is simply <strong>a subset of artificial intelligence</strong>. This tool <strong>gives the systems the ability to learn and improve themselves automatically from experience </strong>without being additionally programmed. One of the most relevant examples of machine learning experimentation is Deep Blue. The AI that defeated the world’s chess champion in 1977, used a method called “tree search algorithms” to evaluate millions of moves at every turn.</p><p>Thus, <strong>ML requires to work with datasets by examining and comparing the data to find common patterns exploring nuances</strong> — this is where the data scientist comes in.<strong> </strong>Another example of using machine learning is in the video prediction systems in platforms like Netflix, YouTube, and Amazon. As a more professional way, <strong>machine learning can also improve productivity of a company by being a part of OCR.</strong></p><h3><strong>The role of Machine learning in OCR.</strong></h3><h4><strong>Machine learning and OCR</strong></h4><p>Before understanding the role of machine learning in OCR, we strongly recommend you to have a quick look at <a href="https://medium.com/@youngapp/top-5-ocr-tools-4e299e38ba72">our previous article on OCR</a>. It explains the basics of OCR and the major companies of this sector.</p><p>To understand how machine learning helps OCR, we need to clarify how OCR works. An optical character recognition is a tool that allows to recognize and select data from any document. You can use <a href="https://youngapp.co/blog/Automated-the-accounting-process">OCR in accounting</a> or it can be also used in the bank industry for <a href="https://youngapp.co/blog/Optimizing-loan-applications-processing">automating the loaning process</a> and <a href="https://youngapp.co/blog/3-ways-of-using-OCR-APIs-to-improve-your-business">much more</a>.</p><p>So, what is the role of machine learning in OCR? Let’s explore it! In order to recognize and classify the different data, the OCR system must be trained. For this, <strong>the system will follow 3 steps: training, testing, validating</strong>. The invoice example can help to understand this operation.</p><h4><strong>Example of Machine learning in OCR</strong></h4><p>In our example, we will analyze the data extracting from an invoice. Just imagine, we want to extract the tax. First, we are going to take a 100 invoices for the <strong>data</strong> <strong>training part</strong>. Then, the data scientist will use <a href="https://www.investopedia.com/terms/n/neuralnetwork.asp">neural networks</a> using deep learning <strong>showing to the systems where to find the “tax”</strong>. <strong>This operation will be repeated</strong> for 80 invoices, the 20 others are going to be used for the next part.</p><p>After teaching the systems where to found the tax, it’s now the turn to the algorithm to work. On the 20 last invoices, <strong>the algorithm mission will be to determine where is the data that it learned</strong> and just like that, we are already at the<strong> data test part</strong>.</p><p>Finally, the last step: <strong>data validation</strong>. As the name says, it’s the part where <strong>the human will confirm the work of the machine</strong>. Indeed, at the end of the data test move, the algorithm will announce an accuracy percentage so we could decide whether it appropriate for us. Let’s say the accuracy is 70%. The data scientist can either choose to accept such accuracy or to continue training and repeat processes till receive a more precise OCR.</p><p>If after reading this article, you would like to know how to apply OCR to your organization, at Young App, we guide companies in their digital transformation adenture with our OCR. We help to improve productivity by digitazing processes. You can first have a look to our <a href="https://youngapp.co/">website</a> or even <a href="https://meetings.hubspot.com/gerard-toko/request-demo">make an appointment</a>. We will be more than happy to be your advisors.</p><h3>Young App contacts</h3><p>Let’s stay connected🙌 Join us:<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5e1254780caf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Top 5 of Artificial Intelligence and Machine learning courses]]></title>
            <link>https://medium.com/@dataleonlabs/top-5-of-artificial-intelligence-and-machine-learning-courses-7be37eece14e?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/7be37eece14e</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[courses-and-training]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Mon, 10 Aug 2020 15:39:22 GMT</pubDate>
            <atom:updated>2020-08-11T11:18:26.382Z</atom:updated>
            <content:encoded><![CDATA[<h3>Top 5 courses on Artificial Intelligence and Machine Learning</h3><h3>Introduction</h3><blockquote>The curiosity in artificial intelligence (AI) is taken to a whole new level these past years. Every day <a href="https://youngapp.co/">new startups</a>, new tools, new innovations are growing. Two words, in the middle of all this interest: <a href="https://towardsdatascience.com/ai-machine-learning-deep-learning-explained-simply-7b553da5b960">machine learning</a> (ML). This term is now always mentioned when we talk about AI. Nowadays, though, people who interested in learning more about this technology won’t have time to go back to college or spend a whole year on a training course. For this reason, we decided to created this article. <strong>We will present our top 5 free AI or ML courses</strong> so you can be sure to have the basic knowledge as well.</blockquote><figure><img alt="Machine learning courses" src="https://cdn-images-1.medium.com/max/1000/0*maSwJ7nC8Z9xSTLq" /></figure><h3>Google AI</h3><p><a href="https://ai.google/education/">Go to courses</a></p><figure><img alt="Free AI &amp; ML courses by Google AI" src="https://cdn-images-1.medium.com/max/1024/1*DCcOJr3sq4kC7feDVAFX3w.png" /><figcaption>Google AI Platform</figcaption></figure><p>As you might understand, these courses are taught by Google and more specifically by some ML <strong>experts from Google</strong>. What we like about this formation is that it is open to all levels. Anybody can improve their skills in that way.</p><p>Moreover, the platform is well-designed. Indeed, you will have three options to fill in “I am”, “type of content” and “stage of ML content”. These three features will help you to find the right course, documentation, video, tutorials, or guide for you.</p><p>If you are still doubtful about Google AI, here are some content you can found:</p><ul><li>“Introduction to machine learning problem framing” (courses)</li><li>“Machine learning crash course with TensorFlow APIs” (courses)</li><li>“Responsible AI practice” (guide)</li><li>“AI adventures: 7 steps of machine learning” (video)</li></ul><h3>Stanford university — Coursera</h3><p><a href="https://www.coursera.org/learn/machine-learning?aid=true#syllabus">Go to courses</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*v4AlIoKsWMjNxagqml159w.png" /><figcaption>Machine learning on Coursera platform</figcaption></figure><p>Coursera is a platform that proposes a<strong> variety of different courses that are taught by universities all over the world</strong>. You can find courses from HEC in Paris, Imperial College London in the United Kingdom, University of Amsterdam in the Netherlands, but also UPenn, Duke, Yale or Stanford in the USA, and much more.</p><p>If you visit the website, you will find many interesting ML courses. Through this article, we agreed to discuss the Stanford university formation. At the end of these courses, <strong>you will know some effective ML techniques but also some of Silicon Valley’s best practices innovative about machine learning and AI</strong>. What we appreciate with this Stanford class, is that you will gain both practical and theoretical knowledge.</p><p>The last point with this formation, know that all the courses are free, however, if you want the certificate, you will have to pay 79$.</p><h3>AI School by Microsoft</h3><p><a href="https://aischool.microsoft.com/en-us/machine-learning/learning-paths">Go to courses</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sovpo1pKuk0JF3Sn0oREcg.png" /><figcaption>AI School website</figcaption></figure><p>AI School is a Microsoft platform, and like Google AI, <strong>it’s available to beginners and experienced people</strong>. More than an intuitive website, what is interesting about AI School by Microsoft, is that you can find short 2-hour courses like “Deep learning explained” as well as education programs.</p><p>For instance, you can do the “object recognition using a custom deep neural network model” formation of 205 hours. However, if any training seems too long, you do have the option to choose which class you want to take. If we are taking back the 205 hours formation, you can opt to follow just the “object detection and classification” course of 2 hours.</p><p>As you can see, AI School is a fruitful platform with multiple free courses on artificial intelligence, deep learning, machine learning, etc.</p><h3>Udacity</h3><p><a href="https://www.udacity.com/school-of-ai">Go to courses</a></p><figure><img alt="Free courses on artificial intelligence" src="https://cdn-images-1.medium.com/max/1024/1*8CzT7L1ScvrJNc4b824P_A.png" /><figcaption>Udacity website</figcaption></figure><p>If you are interested in IT and you want to develop your skill, you certainly have heard about “Udacity”. <strong>This website allows you to learn different programs in IT or deep tech that is divided into 6 different categories</strong>:</p><ul><li>Programming</li><li>Data science</li><li>Artificial intelligence</li><li>Autonomous systems</li><li>Cloud computing</li><li>IT business</li></ul><p>As you can imagine, you will find the machine-learning courses in the AI programs. It contains some classes for beginner, intermediate and advanced learners. This means that you can find courses like “Intro to ML with PyTorch” (beginner); “Deep learning” (intermediate); “Computer vision” (advanced).</p><p>Finally, what we like about this platform, is that <strong>the courses cover a large variety of topics.</strong> Moreover, you have a specific estimation period before you start. For example, with the “Deep Learning” course, you can see “4 months at 12 hours per week.” Before this forecast, you have a segment called “What you will know” that prepares you for this training.</p><h3>Element of AI</h3><p><a href="https://course.elementsofai.com/">Go to courses</a></p><figure><img alt="AI and Machine learning courses for beginners" src="https://cdn-images-1.medium.com/max/1024/1*K-z4Q96vhzALVW5wrJe_fA.png" /><figcaption>Element of AI platform</figcaption></figure><p>In the fifth position, we have Element of AI! This website gives you access to free online courses created by the company Reaktor and the University of Helsinki. <strong>Their purpose was to make AI accessible to everyone</strong>.</p><p>Therefore, the courses available on this platform are mostly for beginners. You will find 6 chapters from “What is AI?” to “AI problem solving” passing by “Machine learning”. Following Element of AI courses, will allow you to acquire a solid foundation on AI.</p><p>What we like about it, <strong>is the user-friendly interface</strong> but also the fact that <strong>their courses combine theoretical knowledge with practical exercises</strong>.</p><p>If you are interested in machine learning, know that at Young App, we are using this process in our <strong>OCR technology</strong>. Our purpose is to help companies to automate workflow by using the OCR solutions. If you would like to develop automation skills, don’t hesitate to visit our <a href="https://youngapp.co/">website</a> or even <a href="https://meetings.hubspot.com/gerard-toko/request-demo">make an appointment</a>. We will be more than happy to be your advisors.</p><h3>Young App contacts</h3><p>Let’s stay connected🙌 Join us:<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7be37eece14e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Automated the accounting process]]></title>
            <link>https://medium.com/@dataleonlabs/automated-the-accounting-process-a65ed42eabf5?source=rss-7c9d738e491f------2</link>
            <guid isPermaLink="false">https://medium.com/p/a65ed42eabf5</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ocr-software]]></category>
            <category><![CDATA[accounting]]></category>
            <dc:creator><![CDATA[Dataleon]]></dc:creator>
            <pubDate>Fri, 24 Jul 2020 13:29:54 GMT</pubDate>
            <atom:updated>2020-08-11T12:05:57.486Z</atom:updated>
            <content:encoded><![CDATA[<h3>Automated accounting process</h3><h4>Case study | Accounting | OCR | AI</h4><h3>Overview</h3><blockquote>An OCR workflow allowing you to win some time on the accounting process by automating extracting data from invoices (bills).</blockquote><figure><img alt="Accounting automated" src="https://cdn-images-1.medium.com/max/1000/0*JD4-2hWCZOZLpvR4" /></figure><h3>Context</h3><p>Accounting and banking are both sectors that require multiple time-consuming tasks. These repeated missions can also impact the productivity of many employees. Nevertheless, today we assist in the rise of tools that allow us to automate all these processes. Indeed, automation technologies such as RPA are expected to have a potential economic impact of <a href="https://www.accountingtoday.com/opinion/the-challenge-for-cpa-firms-applying-rpa-in-the-real-world">6.7 trillion dollars by 2025</a>. Moreover, tools like OCR can also be an alternative to RPA and help you to automate your accounting or banking processes. This is what we will analyze in this use case.</p><p>Before beginning, let’s clarify two words that are going to use a lot in this article: invoice and bill. Invoice is the document issued by us, and the bill is the document issued by another company for us, so we are issuing invoices and receiving bills.</p><h3>Challenge</h3><ul><li>Extract information from invoices (bills) using OCR technology.</li><li>Add possibility to search, edit, and classify invoices (bills).</li></ul><h3>Solution</h3><ul><li>Connect OCR API to company software</li><li>Train software using ML algorithms</li><li>Recognize, collect, and classify information over APIs</li><li>Enable converting data and storing in databases</li></ul><h3>Result</h3><p>At the end of this automated workflow, we can observe that the data are being extracted in seconds by autoscanning and classifying documents. Compliance checks and structuring data are provided so invoices may be quickly reviewed and analyzed.</p><p>Finally, you won’t have to waste hours on entering accounting invoices (bills) to the system thanks to the OCR technology.</p><p>If you decide to automate your accounting, at Young App, we will gladly help your company with all API &amp; OCR projects you may have.</p><h3>Young App contacts</h3><p>Let’s stay connected 🙌<br><a href="https://youngapp.co/"><strong>Website</strong></a><strong><br></strong><a href="https://www.linkedin.com/company/young-app"><strong>LinkedIn</strong></a><strong><br></strong><a href="https://manual.youngapp.co/connectors"><strong>GitBook</strong></a><strong> </strong>(documentation)<br><a href="https://github.com/youngapp/yap"><strong>GitHub</strong></a> (we 🙏❤️appreciate if you could click the ⭐️-like to support us)<br><a href="https://twitter.com/youngapp_pf"><strong>Twitter</strong></a> (🔥hottest news about API, microservices, serverless technologies)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a65ed42eabf5" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>