Day 2 #SheBuildsOnAzure

KEY HIGHLIGHTS

• Circular form of Ai
• +Target audience sensing
• +Pixels RGB
• +ML learning model
• +Clustering
• +UCF recommendation system
• +KNN Algorithm

Circular AI Circular AI, also known as circular artificial intelligence, refers to the integration of AI technologies and principles into circular economy frameworks. The circular economy aims to minimize waste, promote resource efficiency, and foster sustainable production and consumption patterns. Circular AI leverages artificial intelligence techniques and algorithms to enable and enhance the circular economy’s goals and strategies.

• Circular AI can be applied across various domains and sectors to optimize resource utilization, improve waste management, and drive sustainable practices. Here are a few examples of how circular AI can be implemented: + Waste Management: Circular AI can help optimize waste collection and recycling processes by utilizing machine learning algorithms to analyze data from sensors and smart devices. It can identify patterns, optimize routes for waste collection, and provide insights for waste reduction and recycling strategies.
• + Product Lifecycle Management: Circular AI can facilitate the tracking and tracing of products throughout their lifecycle. By integrating AI algorithms with IoT sensors, it becomes possible to collect and analyze data on product usage, performance, and maintenance. This information can be used to optimize product design, enable predictive maintenance, and support efficient reuse or recycling of materials.
• + Supply Chain Optimization: Circular AI can optimize supply chain processes by analyzing data from various sources, such as logistics, manufacturing, and inventory management.
• + Sustainable Energy Systems: Circular AI can help integrate renewable energy sources, such as solar and wind power, into existing energy grids. Circular AI holds significant potential to transform industries and enable a more sustainable and resource-efficient future. By combining AI technologies with circular economy strategies, it becomes possible to optimize processes, minimize waste generation, and promote the reuse and recycling of materials, thereby contributing to a more sustainable and circular society. RGB
• + RGB stands for Red, Green, and Blue, which are the primary colors used in additive color models. In the context of pixels, RGB refers to a color model where each pixel is represented by a combination of these three primary colors.
• + In digital imaging and display systems, an image is composed of a grid of pixels. Each pixel can emit or display a specific color by varying the intensity of its red, green, and blue components. The RGB color model assigns a value between 0 and 255 to each of the red, green, and blue channels, representing the intensity of that color component. By combining different intensities of red, green, and blue, it is possible to create a wide range of colors.
• + The RGB color model follows the additive color mixing principle. When red, green, and blue light are combined at full intensity (255, 255, 255), they create white light. On the other hand, when all color channels are set to their minimum intensity (0, 0, 0), no light is emitted, resulting in black.
• + To represent a specific color, the RGB values of each pixel are usually encoded using 24 bits, with 8 bits allocated to each color channel (8 bits for red, 8 bits for green, and 8 bits for blue). This encoding scheme is commonly known as 24-bit color or True Color, as it can represent over 16 million different colors.
• + In summary, RGB is a color model used to represent colors in digital imaging systems, where each pixel’s color is defined by its intensity of red, green, and blue components. By combining different intensities of these primary colors, a wide spectrum of colors can be achieved. Machine learning (ML) learning models
• + ML models refer to algorithms or computational models that are designed to learn and make predictions or decisions based on data. These models are a fundamental component of machine learning, a subfield of artificial intelligence that focuses on developing algorithms and techniques that enable computers to learn from and make predictions or decisions without being explicitly programmed.
• + There are several types of ML learning models, each with its own characteristics and applications. Here are some commonly used ML learning models:
• + Linear Regression: Linear regression is a supervised learning model used for predicting continuous values. It establishes a linear relationship between the input variables (features) and the target variable by finding the best-fit line that minimizes the sum of squared differences between the predicted and actual values.
• + Logistic Regression: Logistic regression is used for binary classification problems where the target variable has two classes. It estimates the probability of an input belonging to a particular class using a logistic function and assigns it to the most probable class.
• + Decision Trees: Decision trees are versatile supervised learning models that can be used for both classification and regression tasks. They create a tree-like model of decisions and their possible consequences based on the features of the input data.
• + Support Vector Machines (SVM): SVM is a powerful supervised learning model used for both classification and regression tasks. It finds an optimal hyperplane that separates data points of different classes with the maximum margin, allowing for better generalization.
• + Neural Networks: Neural networks are a class of deep learning models inspired by the structure and function of the human brain. They consist of interconnected layers of artificial neurons (nodes) and are capable of learning complex patterns and representations from data. + Naive Bayes: Naive Bayes is a probabilistic classifier based on Bayes’ theorem with an assumption of independence between features. It is particularly useful for text classification and spam filtering tasks.
• + K-Nearest Neighbors (KNN): KNN is a non-parametric model used for both classification and regression tasks. It classifies or predicts a new data point based on the majority vote or average of its k nearest neighbors in the feature space. Clustering
• + Clustering is an unsupervised machine learning technique used to group similar data points together based on their inherent patterns and similarities. The goal of clustering is to identify meaningful groups or clusters in the data without any prior knowledge or labeled examples.
• + In clustering, the algorithm analyzes the input data and assigns each data point to a cluster based on its similarity to other data points. The similarity or dissimilarity between data points is typically measured using distance metrics, such as Euclidean distance or cosine similarity. The clustering algorithm aims to minimize the intra-cluster distance (distance between data points within the same cluster) and maximize the intercluster distance (distance between data points in different clusters).
• + There are various clustering algorithms available, each with its own strengths and assumptions. Here are some commonly used clustering algorithms: + K-Means Clustering: K-means is one of the most popular clustering algorithms. It partitions the data into k clusters, where k is a predefined number chosen by the user. It iteratively assigns data points to clusters and updates the cluster centroids until convergence, aiming to minimize the sum of squared distances between data points and their assigned cluster centroids.
• + Hierarchical Clustering: Hierarchical clustering builds a hierarchy of clusters by iteratively merging or splitting clusters based on their similarity. It does not require specifying the number of clusters in advance and can result in a tree-like structure called a dendrogram, which allows different levels of granularity in cluster assignments. + DBSCAN (Density-Based Spatial Clustering of Applications with Noise): DBSCAN is a density-based clustering algorithm that groups together data points that are closely packed together and separates points in sparse regions. It identifies core points, which have a sufficient number of nearby points, and expands clusters by connecting density-reachable points.
• + Mean Shift Clustering: Mean Shift is an iterative algorithm that seeks to find the modes or high-density regions of the data distribution. It starts with initial points in the feature space and iteratively shifts them towards the regions of higher data density. The final cluster assignments are determined by the convergence of the algorithm.
• + Gaussian Mixture Models (GMM): GMM assumes that the data is generated from a mixture of Gaussian distributions. It aims to es timate the parameters of these Gaussian distributions and assign data points to the most likely cluster based on their probability distribution.
• + Clustering can be used in various applications, including customer segmentation, image segmentation, anomaly detection, document clustering, and many more. It helps to discover hidden patterns, group similar data points together, and gain insights into the structure of the data. The choice of clustering algorithm depends on the nature of the data, the desired interpretability, the number of clusters, and other factors specific to the problem at hand. UCF (User-based Collaborative Filtering)
• + The UCF (User-based Collaborative Filtering) recommendation system is a collaborative filtering approach used to make personalized recommendations based on the preferences and behavior of similar users. It relies on the assumption that users who have similar tastes and preferences in the past will have similar preferences in the future.
• + The UCF recommendation system works in the following steps:
• + User-Item Matrix: The system first creates a user-item matrix, where rows represent users and columns represent items (e.g., movies, products). The matrix contains ratings or preferences given by users to items. If a user has not rated an item, it may be represented as a missing value or zero.
• + Similarity Calculation: The system then calculates the similarity between users based on their ratings. Various similarity metrics can be used, such as cosine similarity or Pearson correlation coefficient. The similarity metric measures how closely the rating patterns of two users match.
• + Nearest Neighbors: Once the similarity between users is calculated, the system identifies the nearest neighbors of a target user. These nearest neighbors are users who have similar preferences to the target user. The number of nearest neighbors to consider can be defined in advance.
• + Rating Prediction: Finally, the system predicts the ratings that the target user would give to items they have not yet rated. It does this by aggregating the ratings of the nearest neighbors for those items. Weighted averages or other techniques can be used to give more weight to the ratings of closer or more similar neighbors.
• + Recommendation Generation: Based on the predicted ratings, the system generates recommendations for the target user. The items with the highest predicted ratings are recommended to the user.
• + The UCF recommendation system has some advantages, such as being able to provide personalized recommendations and not requiring detailed item information. However, it also has limitations, including the “cold start” problem, where new users or items with few ratings may not receive accurate recommendations.
• + It’s worth noting that there are other recommendation techniques as well, such as item-item collaborative filtering, content-based filtering, and hybrid approaches that combine multiple methods to improve recommendation accuracy and coverage. The choice of recommendation system depends on factors such as the available data, system requirements, and the specific characteristics of the recommendation domain. K-nearest Neighbours (KNN) K-nearest neighbors (KNN) comes under supervised learning algorithm thatinvolves grouping objects based on their similarities. It is used for classification and regression tasks. It is a nonparametric algorithm, meaning it does not make assumptions about the underlying data distribution. Instead, it makes predictions based on the proximity or similarity of data points. Imagine of having a set of 5 different colored balls: red, blue, green, yellow, and orange. KNN works by classifying an unknown ball based on the colors of its nearest neighbors. For example, if the majority of the nearest neighbors of an unknown ball are Red, then it is likely to be classified as a Red ball. KNN uses the concept of proximity to make predictions. In this case, the “distance” between colors represents how similar they are. KNN is a simple and intuitive algorithm that can be applied to various scenarios beyond color classification.TheKNN algorithm is relatively easy to understand and implement, making it a popular choice for many classification and regression problems. However, it may not perform well in high-dimensional spaces or with datasets where the majority class dominates. It is always recommended to consider the characteristics of the data and experiment with different values of K to achieve optimal results.