<?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 Hassan Muhammad Adam on Medium]]></title>
        <description><![CDATA[Stories by Hassan Muhammad Adam on Medium]]></description>
        <link>https://medium.com/@hassanmuhammadadam1?source=rss-32ed7b688c7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*NtMEUdbk_mEiWyHWEyJf2g.jpeg</url>
            <title>Stories by Hassan Muhammad Adam on Medium</title>
            <link>https://medium.com/@hassanmuhammadadam1?source=rss-32ed7b688c7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 12:34:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@hassanmuhammadadam1/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[Reinforcement  learning]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/reinforcement-machine-learning-b7e4a3b64f99?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/b7e4a3b64f99</guid>
            <category><![CDATA[reinforcement-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Sat, 15 Jul 2023 22:56:03 GMT</pubDate>
            <atom:updated>2023-07-17T17:22:10.378Z</atom:updated>
            <content:encoded><![CDATA[<h3>Reinforcement learning</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/436/1*LVIe6FiVp7587VKKVp9xPw.png" /></figure><h4>Introduction to Reinforcement learning</h4><p>Machine learning includes the discipline of reinforcement learning. It involves acting appropriately to maximize reward in a certain circumstance. It is used by a variety of programs and machines to determine the optimal course of action to pursue in a given circumstance. There is no correct answer in reinforcement learning, but the reinforcement agent decides what to do to complete the task. This is different from supervised learning, where the training data includes the answer key and the model is trained with that answer. It is obligated to gain knowledge from its experience in the absence of a training dataset.</p><p>The study of decision-making is called reinforcement learning (RL). It involves understanding how to act in a situation to reap the most benefits. Data for RL is gathered from machine learning systems that employ a trial-and-error process. Input for either supervised or unsupervised machine learning does not include data.</p><h4>Difference between Reinforcement learning and Supervised learning:</h4><p>Making decisions consecutively is the foundation of reinforcement learning. Simply put, the next input depends on the output of the previous input, and the output depends on the state of the current input.</p><p>The initial input, or the input provided at the beginning, is considered when making decisions in supervised learning.</p><h4>Reinforcement learning components</h4><p>Reinforcement learning components are as follows:</p><ol><li>Policy</li><li>Reward function</li><li>Value function</li><li>Model of the environment</li></ol><p><strong>Policy:</strong> A policy outlines the behavior of learning agents for a specific time frame. It is a mapping between perceived environmental states and the activities that should be conducted in such states.</p><p><strong>Reward function:</strong> In a reinforcement learning issue, the goal is defined by the reward function. A reward function is a function that calculates a score based on how the environment is doing.</p><p>Value functions define what is beneficial over the long run. The total quantity of rewards an agent might anticipate accumulating over the course of their existence, starting from a certain state, is the value of that state.</p><p><strong>Model of the environment:</strong> Planning often involves using models.</p><p><strong>Application of Reinforcement Learning</strong></p><p>Robotics: In organized environments where the task is repetitive in nature, such as the assembly line of an automobile manufacturing plant, robots with pre-programmed behavior are useful.</p><p>An expert chess player moves. Planning and preparing for potential responses and counterreplies inform the decision.</p><p>Real-time parameter changes are made via an adaptive controller in a petroleum refinery.</p><p><strong>Advantages of Reinforcement learning</strong></p><p>Reinforcement learning can be used to tackle extremely difficult issues that are intractable using traditional methods.</p><p>The model has the ability to fix mistakes made during training.</p><p>Numerous issues, including those involving decision-making, control, and optimization, can be resolved via reinforcement learning.</p><p><strong>Disadvantages of Reinforcement learning</strong></p><p>1. Using reinforcement learning to solve straightforward issues is not recommended.</p><p>2. The process of reinforcement learning requires a lot of data and computation.</p><p>3. It might be challenging to comprehend and troubleshoot reinforcement learning. It might be challenging to identify and resolve issues when the agent’s actions are not always obvious.</p><p>Conclusion: We have seen the concept of reinforcement learning. And also, its applications, and so on. You can further your reading by visiting the following link or other necessary material that will expand your understanding of it.</p><p><a href="https://github.com/arewadataScience/ML-4-Beginners-Arewa-Datascience/tree/main/8-Reinforcement">https://github.com/arewadataScience/ML-4-Beginners-Arewa-Datascience/tree/main/8-Reinforcement</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b7e4a3b64f99" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TIME SERIES IN MACHINE LEARNING]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/time-series-in-machine-learning-f0a9f001ade1?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/f0a9f001ade1</guid>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Fri, 07 Jul 2023 08:47:03 GMT</pubDate>
            <atom:updated>2023-07-07T09:03:04.041Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/743/1*yUFyQVDC40zdPT-DaCFBpQ.jpeg" /></figure><h3><strong>Introduction to Time Series</strong></h3><p>Time series forecasting is the process of making future value predictions using data from the past. Weather, sales, and stock price predictions are a few examples from various businesses. It has been used more recently to forecast price movements for cryptocurrencies like Bitcoin and Ethereum. Every data scientist should have some understanding of the methodologies that can be used to perform time series forecasting given the prominence of time series forecasting applications in many different domains.</p><p>For time series forecasting, a variety of techniques are available. The statistical model known as Autoregressive Moving Average (ARMA), which predicts future values based on past values, is one of the most often utilized. But because it misses seasonal trends, this approach to time series prediction is incorrect. Additionally, it is predicated on the idea that the time series data is stationary, i.e., that its statistical characteristics would not alter over time. But because ARMA is based on an idealized assumption that doesn’t hold in reality, the results could be biased.</p><p>Python provides many easy-to-use libraries and tools for performing time series forecasting in Python.</p><h3>Characteristics of Time Series Forecasting</h3><p>You can find trends and get more accurate predictions by researching a series&#39; previous behavior. Many time series show one or more of the following characteristics when plotted:</p><p>.Trends</p><p>.Seasonal and unseasonal cycles</p><p>.Pulse and steps</p><p>.Outliers</p><p><strong>1.Trends</strong>: A trend is an incremental change in the series&#39; level or a tendency for the values of the series to rise or fall over time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/528/1*slu3VPkpreHvinSMreksiQ.jpeg" /></figure><p><strong>2.Seasonal cycle:</strong>The interval of your series is connected to seasonal cycles. For instance, quarterly and annual data cycles are normal for monthly data. An important yearly cycle with a high every December or a big quarterly cycle with a low in the first quarter may be visible in a monthly series. Seasonality is the display of a seasonal cycle in a series.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*0P-MVhh6-v9LJDsK9tvY_w.jpeg" /></figure><p><strong>3.pulse and Steps:</strong>Many series experience abrupt changes in level.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/526/1*4QNCCHhlahhQ6qxZJS_JkQ.jpeg" /></figure><p><strong>4.Outliers</strong>:</p><p>Outliers are changes in a time series&#39; level that cannot be accounted for. These data deviate significantly from the rest of the series, which can have a significant impact on the analysis and, as a result, the time series model’s capacity for forecasting.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/803/1*3Lcms2X19Lj6cBrgeAaJyA.jpeg" /></figure><h3>Application Of Time Series</h3><p>-Retail Sales</p><p>-Stock Prices</p><p>-Weather</p><p>-Heart Rate</p><p>-Subscribers</p><p>Conclusion</p><p>Time series forecasting are essential part of your ours , but for further reading and understanding of Time series visit <a href="https://github.com/arewadataScience/ML-4-Beginners-Arewa-Datascience">https://github.com/arewadataScience/ML-4-Beginners-Arewa-Datascience</a></p><p><a href="http://kaggle.com/">kaggle.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f0a9f001ade1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Sentimental Analysis]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/introduction-to-sentimental-analysis-55e332364c80?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/55e332364c80</guid>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Sun, 25 Jun 2023 17:03:24 GMT</pubDate>
            <atom:updated>2023-07-08T12:49:13.243Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*mI9zMDlPkU_v92defRaFeg.jpeg" /></figure><p>A common task in natural language processing is sentiment analysis. The objective of sentiment analysis is to categorize the text according to the attitude or mood it conveys, which may be positive, negative, or neutral.</p><p>Sentiment Assessment<br>The method of determining whether a block of text is good, negative, or neutral is known as sentiment analysis. Sentiment analysis seeks to understand public sentiment in order to analyze it for business purposes. It emphasizes emotions, such as happiness, sadness, and anger, in addition to polarity (positive, negative, and neutral). It makes use of a variety of Natural Language Processing techniques, including Automatic, Hybrid, and Rule-based.</p><p>Sentiment analysis is another task that machine learning excels at. Finding &#39;good&#39; and &#39;negative&#39; words and phrases is one non-ML method for analyzing sentiment. Next, determine the overall sentiment of a fresh text passage by adding up the positive, negative, and neutral terms.</p><h4>Sentiment analysis techniques</h4><p><strong>Fine-grained sentiment analysis:</strong>The polarity base affects this. Very positive, positive, neutral, negative, or very negative are all possible design options for this category. On a scale of 1 to 5, the evaluation is done. If the rating is 5, it is highly positive; if it is 2, it is negative; and if it is 3, it is neutral.</p><p><strong>Emotion detection:</strong> encompasses feelings such as joy, happiness, sadness, rage, upset, merriment, and so forth. The lexical approach of sentiment analysis is another name for it.</p><p><strong>Aspect-based sentiment analysis: element</strong>-based analysis is employed because it concentrates on a certain element. For example, if someone wants to evaluate a cell phone’s feature, they would look at aspects like the battery, screen, and camera quality.</p><p><strong>Multilingual sentiment analysis:</strong>Multiple languages must be categorized as positive, negative, or neutral when they are spoken together. This is quite demanding and challenging in comparison.</p><p>Working with sentimental Analysis Using python Libraries.</p><ol><li>Text blob : A Python package for natural language processing is called Text Blob. It’s quite easy to perform sentiment analysis with Text Blob.</li></ol><p>Installation</p><pre>pip install textblob<br></pre><p>Importing Text Blob:</p><pre>from textblob import TextBlob<br><br></pre><p>2. Vader : A rule-based sentiment analyzer called VADER (Valence Aware Dictionary and Sentiment Reasoner) has been trained on social media text. Similar to Text Blob, its use in Python is rather straightforward.</p><p>Installation</p><pre>pip install vaderSentiment<br><br></pre><p>Importing</p><pre>from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer<br></pre><h3>Applications</h3><p><strong>Social media:</strong> Take comments on Instagram as an example. Here, all reviews are examined and divided into three categories: favorable, negative, and neutral.</p><p><strong>Customer service:</strong> Sentiment analysis techniques are used to complete all comments in the Play Store that are rated from 1 to 5.</p><p><strong>Marketing Sector:</strong> In the marketing field, a certain product must be evaluated to determine whether it is good or terrible.</p><p><strong>Reviewer side</strong>: Each reviewer will read the comments, check their accuracy, and provide an overall assessment of the product.</p><h4>Conclusion</h4><p>Sentiment analysis is very important in natural processing.</p><p>and in the world at large. you can find helpful resources through the below link, <a href="https://github.com/arewadataScience">https://github.com/arewadataScience</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=55e332364c80" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A description of natural language processing]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/a-description-of-natural-language-processing-8f47c1cf49?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/8f47c1cf49</guid>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Sat, 24 Jun 2023 22:02:08 GMT</pubDate>
            <atom:updated>2023-07-08T13:13:39.461Z</atom:updated>
            <content:encoded><![CDATA[<h3>Natural language processing In Machine Learning</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/764/1*-PWj1LuMCNo5fc-VTgAO7A.jpeg" /></figure><p>The application of artificial intelligence (AI) to the interaction of computers and human languages is known as natural language processing (NLP). To analyze, comprehend, and produce natural language text and speech, NLP is employed. The objective of NLP is to make it possible for computers to comprehend and interpret human language in a manner that is comparable to how humans do so.</p><p>The term &quot;natural processing language,&quot; or &quot;NLP,&quot; refers to one of the well-known applications of machine learning that is currently being employed worldwide.</p><p>A branch of computer science and artificial intelligence called “natural language processing” (NLP) studies how computers and human languages interact. NLP’s main objective is to make it possible for computers to comprehend, interpret, and produce natural language in a manner similar to that of humans.</p><p>Computational linguistics, machine learning, and statistical modeling are a few of the approaches used in NLP. These methods are applied to the analysis, comprehension, and manipulation of text, speech, and other types of communication that contain human language.</p><p>Language translation, speech recognition, sentiment analysis, text categorization, and information retrieval are a few of the key uses for NLP. Numerous sectors, including finance, healthcare, education, and entertainment, among others, employ NLP.</p><p>The study of natural language processing (NLP), a branch of computer science and artificial intelligence, aims to improve communication between machines and people. Analyzing, comprehending, and producing text and speech-based human language data are all part of it.</p><p>Sentiment analysis, machine translation, text summarization, chatbots, and other uses for NLP are just a few examples. The following are some typical NLP tasks:</p><p>Text classification, also known as spam filtering, sentiment analysis, and topic modeling, is the process of categorizing texts according to their content.</p><p>Recognizing and classifying named entities in text, such as persons, companies, and locations, is known as named entity recognition (NER).</p><p>Speech component (POS) Assigning a noun, verb, adjective, or adverb to each word in a sentence is known as tagging.</p><p>Analysis of a text’s sentiment, such as whether it is positive, negative, or neutral.</p><p>Text that has been translated automatically from one language into another.</p><h3>What is required for you to begin working with NLP?</h3><p>Python, Visual Studio Code with the extension, and the Python IDE<br>A Python text-processing module called Textblob is available.</p><h3>Applications of Natural processing language</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/554/1*gzaUiWLtD4jZpWwbpF19ig.jpeg" /></figure><ul><li>NLP approaches are applied in a variety of contexts, such as:<br>NLP techniques are used to convert speech to text, which is important for activities like dictation and voice-activated assistants. Speech recognition and transcription.</li><li><strong>Language translation:</strong> Text can be translated from one language to another using NLP techniques, which is helpful for e-commerce and other tasks that require international communication.</li><li><strong>Text summarizing:</strong> NLP techniques are used to condense lengthy text documents into manageable chunks, which is helpful for tasks like document indexing and news summaries.</li></ul><p>And so on.</p><p>-Document Analyser</p><p>-Sentiment Analysis</p><p>-Chatbots</p><p>-Email Filtering</p><p>-Automatic Summarization</p><p>-Online Searching</p><p>-Social media monitoring</p><p>-Email Assistant</p><h3>Conclusion</h3><p>We have seen natural processing language and it’s application you can further reading using the below links. <a href="https://github.com/arewadataScience">https://github.com/arewadataScience</a></p><p><a href="https://scikit-learn.org/stable/">https://scikit-learn.org/stable/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8f47c1cf49" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[INTRODUCTION TO CLUSTERING]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/introduction-to-clustering-86ab9938cddd?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/86ab9938cddd</guid>
            <category><![CDATA[clustering-algorithm]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Sun, 11 Jun 2023 19:11:46 GMT</pubDate>
            <atom:updated>2023-07-08T13:37:36.253Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/680/1*VFMtgjSWOpS94RK-dSmijg.jpeg" /></figure><h3>Introduction To Clustering</h3><p>In essence, it is a kind of unsupervised learning technique. The process of drawing references from datasets of input data without labeled replies is known as unsupervised learning. It is typically used as a method to identify the groups, generative qualities, and significant structures that are inherent in a set of instances.</p><p>The objective of clustering is to divide the population or set of data points into a number of groups so that the data points within each group are more similar to one another and different from the data points within the other groups. It is essentially a grouping of objects based on how similar and unlike they are to one another.</p><p>For instance The clustered data points in the graph below can be categorized into a single group. In the image below, the clusters can be distinguished, and we can count three clusters.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X8gDNl-I9Lcj_48CsYVtyA.jpeg" /></figure><h3><strong>Clustering Methods:</strong></h3><p><strong>Density-Based Methods: </strong>According to these methodologies, clusters are a dense portion of the space with some similarities to and differences from the lower dense region. These techniques can combine two clusters and have good accuracy. Examples include OPTICS (Ordering Points to Identify Clustering Structure), DBSCAN (Density-Based Spatial Clustering of Applications with Noise), etc.</p><p><strong>Hierarchical Based Methods: </strong>Based on the hierarchy, the clusters created using this method create a structure resembling a tree. The already formed cluster is used to create new ones. There are two categories. <br>Agglomerative (from the bottom up)<br>(Top-down strategy) Divisive</p><h3>Clustering Algorithm</h3><p>The simplest unsupervised learning approach to handling the clustering problem is the K-means clustering algorithm. The K-means method divides n observations into k clusters, with each observation belonging to the cluster represented by the nearest mean.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iOPVU49imNwtZGT9dSP3JQ.jpeg" /></figure><p><strong>getting started with clustering</strong></p><p>k-means Affinity propagation<br>Mean shift<br>Special clustering<br>ward hierarchical clustering</p><p><strong>common terms use in clustering</strong></p><p><strong>-</strong>Transductive inference<br>-Inductive inference<br>-flat geometry<br>-non-flat geometry<br>-Euclidean distance<br>-non-euclidean distance<br>-Density</p><p><strong>Transductive inference</strong> is applied to specific test cases.<br><strong>Inductive inference</strong>: is applied to general rules, which are then applied to test cases.<br>flat geometry<br><strong>Euclidean distance</strong>: are measured as the length of a line segment between two points in clusters.<br><strong>non-euclidean distance</strong>:are measured along a curve.<br>Density:Data that is noisy is considered to be noise ‘dense’ It can be more or less dense.</p><h4><strong>Clustering Algorithms</strong></h4><p>hierarchical clustering<br>centroid clustering<br>distribution clustering<br>grid-based clustering</p><h3><strong>Applications of Clustering in different fields:</strong></h3><p>For marketing purposes, it can be used to define and identify consumer segments.</p><p><strong>Biology</strong>: It can be applied to the classification of various plant and animal species.</p><p><strong>Libraries</strong>: They are used to group books together according to themes and content.</p><p><strong>Insurance</strong>: It is used to recognize clients, their policies, and fraud.</p><p>Clustering is a technique used in social network research to locate communities or groups, which can be useful in determining trends, influence, and social behavior.</p><p><strong>Cybersecurity</strong>: Clustering groups similar network traffic or system behavior patterns in order to identify and stop cyberattacks.</p><p><strong>Climate analysis:</strong> Similar patterns of climate data, such as temperature, precipitation, and wind, are grouped using a technique called clustering. This can help us understand how climate change is affecting the environment.</p><p><strong>Crime analysis:</strong> Clustering groups comparable patterns of crime data, such as time, place, and kind, and can be used to pinpoint crime hotspots, forecast future trends in crime, and enhance crime prevention tactics.</p><p>Conclusion:</p><p>Clustering has many applications in machine learning and around the world. Read the following links to explore more about clustering <a href="https://scikit-learn.org/stable/">https://scikit-learn.org/stable/</a></p><p><a href="https://github.com/arewadataScience">https://github.com/arewadataScience</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=86ab9938cddd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[CLASSIFICATION WITH MACHINE LEARNING]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/classification-with-machine-learning-bc7d120b2cc?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/bc7d120b2cc</guid>
            <category><![CDATA[unsupervised-learning]]></category>
            <category><![CDATA[supervised-learning]]></category>
            <category><![CDATA[classification-algorithms]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Fri, 26 May 2023 11:50:48 GMT</pubDate>
            <atom:updated>2023-07-08T14:00:14.827Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/653/1*fqC3rFPC_bRI5-4gtOptSg.jpeg" /></figure><h3>Introduction to Classification</h3><p>Classification is one of the fundamental activities of machine learning.</p><p>It can also be seen as the process of recognition, understanding.</p><p>The practice of classifying information or items into predetermined classes or categories based on their characteristics or traits is known as classification. Classification is a form of supervised learning technique used in machine learning where an algorithm is trained on a labeled dataset to anticipate the class or category of new, untainted data.</p><p>Building a model that can precisely assign a label or category to a new observation based on its properties is the basic goal of categorization. For instance, a classification model might be trained on a dataset of photographs that have been classified as either dogs or cats, and then used to forecast the class of brand-new, unlabeled images of dogs or cats based on attributes like color, texture, and shape.</p><h3>Types of Classification</h3><p><strong>Binary Classification: </strong>The objective of binary classification is to divide the input into one of two classes or categories. Example: We must assess whether a person has a specific disease based on the health conditions that have been provided.</p><p><strong>Multiclass Classification</strong>: The objective of multi-class classification is to assign the input to one of a number of classes or groups. For instance, we need to identify which species of flower our observation belongs to based on information about several flower species.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mI603iWjHqWnw_X2S9I4Xg.png" /></figure><h3>Types of classification algorithms</h3><p><strong>Linear Classifiers</strong>: A linear decision boundary between classes is produced using linear models. They are easy to use and effectively use computing. Here are a few examples of linear classification models:</p><p><a href="https://www.geeksforgeeks.org/understanding-logistic-regression/">https://www.geeksforgeeks.org/understanding-logistic-regression/</a></p><p><strong>Non-linear Classifiers: </strong>A non-linear decision boundary between classes is produced by non-linear models, or non-linear classifiers. They are capable of capturing more intricate connections between the target variable and the input features. Here are a few examples of non-linear classification models:</p><p><a href="https://www.geeksforgeeks.org/k-nearest-neighbours/">https://www.geeksforgeeks.org/k-nearest-neighbours/</a></p><h3>Classification Techniques</h3><ul><li>linear modules</li><li>support vector machine</li><li>Decision trees</li><li>Gaussian processes</li></ul><h3>Types of Machine Learning Classification</h3><h4>Supervised learning Approach</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/296/1*ghK9uVFfXneVpQk-eplNYQ.jpeg" /></figure><ul><li>Regression</li><li>Classification</li></ul><h4>Unsupervised learning Approach</h4><ul><li>Clustering</li><li>Dimensionality reduction</li></ul><h3>Applications of Classification Algorithm</h3><p>Classification algorithms are widely used in many real-world applications across various domains, including:</p><ul><li>Email spam filtering</li><li>Credit risk assessment</li><li>Medical diagnosis</li><li>Image classification</li><li>Sentiment analysis.</li><li>Fraud detection</li><li>Quality control</li><li>Recommendation systems</li></ul><p>Conclusion:</p><p>Classification are very important in machine learning you can read further via the links below. <a href="https://scikit-learn.org/stable/">https://scikit-learn.org/stable/</a></p><p><a href="https://github.com/arewadataScience">https://github.com/arewadataScience</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bc7d120b2cc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[REGRESSION WITH MACHINE LEARNING]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/regression-with-machine-learning-1fef2371b809?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/1fef2371b809</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[scikit-learn]]></category>
            <category><![CDATA[regression]]></category>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Mon, 22 May 2023 23:02:05 GMT</pubDate>
            <atom:updated>2023-07-08T14:16:31.271Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*i-PA4q_hJI5CWIq8zTCzgQ.png" /></figure><h3>Introduction to Regression</h3><p>A supervised machine learning approach known as “linear regression” determines the linear relationship between a dependent variable and one or more independent features. Univariate linear regression occurs when there is only one independent feature; multivariate linear regression occurs when there are multiple independent features. The algorithm’s objective is to identify the optimum linear equation that, given the independent variables, can forecast the value of the dependent variable. The relationship between the dependent and independent variables is shown by the equation as a straight line. The slope of the line shows how much the dependent variable changes when the independent variable(s) are changed by a unit.</p><p>Finance, economics, and psychology are just a few of the disciplines that employ linear regression to analyze and forecast the behavior of a given variable. For instance, linear regression can be used in the field of finance to comprehend the connection between a company’s stock price and its earnings or to forecast the value of a currency based on its historical performance.</p><h3>Build the Linear Regression model from Scratch</h3><h4>Import the necessary libraries:</h4><pre>import pandas as pd<br>import numpy as np<br>import matplotlib.pyplot as plt<br>import matplotlib.axes as ax</pre><h4>Load the dataset and separate input and Target variables</h4><pre>data = pd.read_csv(&#39;data_for_lr.csv&#39;)<br><br>Drop the missing values<br>data = data.dropna()<br><br># training dataset and labels<br>train_input = np.array(data.x[0:500]).reshape(500,1)<br>train_output = np.array(data.y[0:500]).reshape(500,1)<br><br># valid dataset and labels<br>test_input = np.array(data.x[500:700]).reshape(199,1)<br>test_output = np.array(data.y[500:700]).reshape(199,1)</pre><h4>Build the Linear Regression Model</h4><pre>#Prediction on test data<br>y_pred = test_input*parameters[&#39;m&#39;] + parameters[&#39;c&#39;]<br><br># Plot the regression line with actual data pointa<br>plt.plot(test_input, test_output, &#39;+&#39;, label=&#39;Actual values&#39;)<br>plt.plot(test_input, y_pred, label=&#39;Predicted values&#39;)<br>plt.xlabel(&#39;Test input&#39;)<br>plt.ylabel(&#39;Test Output or Predicted output&#39;)<br>plt.legend()<br>plt.show()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/571/1*5cA1-LBwiOMsBZmDHf8iRw.png" /></figure><h3>Environment and tools</h3><ul><li>python</li><li>notebook.ipynb</li><li>vscode</li><li>import dataset</li><li>scikit-learn</li></ul><h3>libraries</h3><ul><li>Matplotlib : graphical tool</li><li>Numpy : Handling numeric data</li><li>Pandas : Analizing and manipulating data</li></ul><h3>Linear regression</h3><h4>How it works</h4><p>Relationship between x and y e.g y = b + mx</p><h3>polynomial regression</h3><h4>How it works</h4><p>Type of linear regression where y is modeled as any nth degree polynomial of x e.g y = b + m1x¹ + m2x² + mnx^n</p><h3>Logistics Regression</h3><ul><li>Binary classification.</li><li>Multinomial classification.</li><li>Ordinal classification.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1fef2371b809" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VISUALIZATION IN MACHINE LEARNING]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/visualization-in-machine-learning-9b85345f8fb3?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/9b85345f8fb3</guid>
            <category><![CDATA[scatter-plots]]></category>
            <category><![CDATA[histograms]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[pie-charts]]></category>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Sun, 14 May 2023 15:43:05 GMT</pubDate>
            <atom:updated>2023-06-30T07:39:32.511Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/290/1*9xe04KIH0bRtG6aYJuooyA.png" /></figure><p>Data visualization is <strong>the representation of data through the use of common graphics, such as charts, plots, infographics, and even animations</strong>. Data visualization is very important in the data science journey, and Matplotlib is a comprehensive Python library for static , animated, and interactive visualization.</p><h3>visualizing quantities</h3><p>The first thing to do when a data set is given is check the amount of data in the set using line charts, scatter plots, and bar plots, which are used in visualizing data quantity.</p><ul><li>Line plots</li><li>Scatter plots</li><li>Bar plots</li></ul><h3>Visualizing Data Distribution</h3><p>Visualizing data distribution can give different insights, such as the frequency of data along the x and y axes. Data distribution can be visualized using histograms, which use numeric data.</p><ul><li>Histograms</li></ul><h3>Visualizing Data Proportion</h3><p>The next thing is data proportion, which is used to compare the percentages of one another. Data proportions can be visualized using pie charts, donut charts, and waffle charts.</p><ul><li>Pie Charts</li><li>Donut Charts</li><li>Waffle Charts</li></ul><h3>Visualizing Relationship</h3><p>Matplotlib helps us compare relationships or similarities between data. We can make data visualization relationships using scatter plots, line plots, and facets grids.</p><ul><li>Scatter Plot.</li><li>Line Charts.</li><li>Facet Grids.</li></ul><h3>Making meaningful visualizations</h3><p>Now, I want to look into making meaningful visualization, i.e., how do we choose the right chart.</p><ul><li>Trends over time to time: line charts</li><li>Compare categories: bar plot, pie chart.</li><li>Relationships: Scatter plots, line charts and facet</li><li>Proportions: pie charts, donut chart and waffle charts</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9b85345f8fb3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Working With Data and Data Preparation]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/working-with-data-and-data-preparation-fc66d1583675?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/fc66d1583675</guid>
            <category><![CDATA[matplotlib]]></category>
            <category><![CDATA[pandas]]></category>
            <category><![CDATA[numpy]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Sun, 14 May 2023 13:26:03 GMT</pubDate>
            <atom:updated>2023-06-30T07:40:18.163Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/820/1*w2XNo6MO_dQ5uLYYDX-YpQ.png" /></figure><p>Python is one of the most widely used programming languages in data science and web development because it contains many libraries and frameworks. The most popular Python libraries used for data science are Pandas, <strong>Numpy</strong>, <strong>SciPy</strong>, and <strong>Matplotlib</strong>.</p><h3>Data comes in three forms</h3><ol><li>Tabular</li><li>Text</li><li>Images</li></ol><h4>Tabular</h4><p><strong>numpy: </strong>is a library for working with tensors</p><p><strong>Pandas</strong> helps you manipulate data frames.</p><p><strong>Matplotlib: a </strong>library for visualization.</p><p><strong>scipy:</strong> library for scientific functions.</p><h4>Image</h4><ul><li>object detection</li><li>Face detection</li><li>Image classification</li></ul><h4>Text</h4><ul><li>Extract keyword</li><li>Extract information</li></ul><h3>Import Libraries</h3><pre>import numpy as np<br>import pandas as pd<br>import matplotlib.pyplot as plt<br>from scipy import ... # you need to specify exact sub-packages that you need</pre><h4>Getting Data</h4><pre>df = pd.read_csv(&#39;file.csv&#39;)</pre><h3>Data Preparation</h3><p>After data collection, the next step is data preparation, which is the most important in data science.</p><h4>The importance of cleaning data</h4><ul><li>Easy to use</li><li>consistency</li></ul><h4>Common Cleaning Goals and Strategies</h4><ul><li>Exploring data set</li><li>Formatting</li><li>Duplication</li><li>Missing Data</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fc66d1583675" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction and fairness in machine learning]]></title>
            <link>https://medium.com/@hassanmuhammadadam1/introduction-and-fairness-in-machine-learning-5afe13b52212?source=rss-32ed7b688c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/5afe13b52212</guid>
            <dc:creator><![CDATA[Hassan Muhammad Adam]]></dc:creator>
            <pubDate>Fri, 31 Mar 2023 22:16:56 GMT</pubDate>
            <atom:updated>2023-06-29T14:33:43.607Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/727/1*zN9KdSaUXjFmO7I1BgQ8jw.jpeg" /></figure><blockquote>Machine learning “is the field of study that gives computers the ability to learn without being explicitly programmed “ Arthur samuel(1959).</blockquote><h4>Related terms confused with machine learning</h4><ol><li>Artificaial Intelligence</li><li>Deep learning.</li></ol><h4>Fairness in machine learning</h4><p>Fairness has to do with giving similar work without discrimination. Gender (male or female), region, and country when using machine learning.</p><h4>Below are the list of principles guides used by @microsoft in AI development.</h4><p>1.Fairness</p><p>2.Reliability and safety.</p><p>3.privacy and security.</p><p>4.Inclusiveness.</p><p>5.Transparency.</p><p>6.Accountability</p><h4>Fairness Harms in machine learning</h4><ul><li>Allocation.</li><li>Quality of service.</li></ul><h4>APPLICATION OF MACHINE LEARNING</h4><p>1.SMART PHONES</p><p>2.CONNECTED DEVICES</p><h4>WHERE MACHINE LEARNING ARE APPLIED</h4><p>1.To predict weather event</p><p>2.To find fake news or propaganda</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5afe13b52212" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>