Description:FIELD OF THE INVENTION
[0001] The present invention relates to the field of Education and Machine Learning, and more particularly, the present invention relates to the creation system for experimentation, creation, and machine learning.
BACKGROUND FOR THE INVENTION:
[0002] The following discussion of the background to the invention is intended to facilitate an understanding of the present invention. However, it should be appreciated that the discussion is not an acknowledgment or admission that any of the material referred to was published, known, or part of the common general knowledge in any jurisdiction as of the priority date of the application. The details provided herein the background if belongs to any publication is taken only as a reference for describing the problems, in general terminologies or principles or both of science and technology in the associated prior art.
[0003] In the digital era, lots of data are being created on daily basis in almost every domain like health industry, agriculture industry, academic, entertainment, social media etc. Based on these data if a trend or pattern can be found or predicted then the resource can be balanced and optimized with increase growth of finance. Usually, to do this, A person has to be sound technically specially in coding for building a machine learning model along with complicated machine learning tools. This makes it hard for people without a technical background to use the power of machine learning.
[0004] JP6923676B2: This innovation contributes significantly to the field of machine learning in a wide sense. To be more specific, it focuses on technologies and platforms that are meant to do machine learning tasks directly on a device. These tasks include training models, making predictions, gathering use data, and executing other ML operations. These tasks are performed without relying on external servers or cloud services.
[0005] US20230334368A1: In general, the present disclosure pertains to a platform that is integrated with machine learning. The machine learning platform has the capability to transform machine learning models that have distinct schemas into machine learning models that have a similar schema. Additionally, it can organize the machine learning models into model groups depending on particular criteria and do pre-deployment evaluations of the machine learning models. It is possible to analyze or employ the machine learning models that are contained within a model group either individually or collectively. A model group and a selector can be deployed in a production environment using the machine learning platform. The selector can learn to dynamically select the model(s) from the model group in the production environment in different contexts or for different input data, based on a score that is determined using certain scoring metrics, such as certain business goals. This can be done in a variety of scenarios.
[0006] JP7252286B2: It is often related to machine learning that this disclosure is being made. With a greater degree of specificity, this disclosure pertains to on-device machine learning platforms and related technologies that make it possible for on-device prediction, training, example gathering, and/or other machine learning tasks or capabilities.
[0007] In light of the foregoing, there is a need for the Creation system for experimentation, creation, and machine learning that overcomes problems prevalent in the prior art. It is like a smart helper that builds machine learning models without writing a single line of code. Think of it like a cooking assistant: bring the ingredients (dataset), and Creation Ground prepares the dish (the model) by automatically washing, chopping, mixing, and cooking—step by step.
OBJECTS OF THE INVENTION:
[0008] Some of the objects of the present disclosure, which at least one embodiment herein satisfies, are as follows.
[0009] The principal object of the present invention is to overcome the disadvantages of the prior art by providing the Creation system for experimentation, creation, and machine learning.
[0010] Another object of the present invention is to provide the Creation system for experimentation, creation, and machine learning that cleans the dataset, wherein if the data is messy or has missing values, the system fixes it automatically.
[0011] Another object of the present invention is to provide the Creation system for experimentation, creation, and machine learning that chooses the important parameters, wherein the system figures out which parameters of the dataset are most useful for making accurate predictions.
[0012] Another object of the present invention is to provide the Creation system for experimentation, creation, and machine learning that trains the model; wherein the system selects the machine learning algorithms and train them by providing the dataset.
[0013] Another object of the present invention is to provide the creation system for experimentation, creation, and machine learning that tries different settings to improve accuracy, just like adjusting seasoning in a recipe
[0014] Other objects and advantages of the present disclosure will be more apparent from the following description, which is not intended to limit the scope of the present disclosure.
SUMMARY OF THE INVENTION:
[0015] The present invention provides a creation system for experimentation, creation, and machine learning.
[0016] System is designed to solve this problem. It is a no-code platform that automates the entire machine learning process. Anyone can upload their data (in formats like CSV or Excel), and the platform handles everything else automatically—without requiring programming.
[0017] System performs:
- Data cleaning (handling missing values and converting categories into numbers)
- Feature selection (picking the most useful parts of the data)
- Model selection (choosing the best machine learning technique)
- Tuning (adjusting the model for the best performance)
- Evaluation (showing how accurate and reliable the model is)
- Deployment (exporting or using the model easily)
[0018] For example, a researcher studying plant growth may have data on rainfall, soil type, and temperature. Without any coding, they can upload the data to System. The platform will process it, find patterns, and build a model that can predict plant yield. This saves time, reduces effort, and avoids the need for technical skills.
[0019] It uses reliable tools in the background (like Scikit-learn) but hides the complexity from the user. The platform works through a user-friendly interface where users only need to follow clear steps. It empowers non-programmers to take advantage of modern technology without needing to learn how to code.
[0020] The most unique part of this invention—System—is that it brings the complete machine learning process into one single, easy-to-use platform that requires no coding at all.
[0021] End-to-End Automation in One Platform: Most platforms automate only parts of the machine learning pipeline. System combines all the key stages—data preprocessing, feature selection, model training, hyperparameter tuning, evaluation, and deployment—in one place. It does this automatically, behind the scenes.
[0022] No Technical Skills Needed: Even users with zero programming knowledge can use it. The platform is designed to be simple enough for students, teachers, small business owners, and researchers with no tech background.
[0023] One-Click Deployment: After building the model, users can immediately use or download it with just one click. There’s no need to understand technical deployment methods or server setups.
[0024] Smart Model & Feature Selection: System doesn’t just build any model—it tests different models and features behind the scenes to choose the most accurate one for your data, without asking the user to decide.
BRIEF DESCRIPTION OF DRAWINGS:
[0025] Reference will be made to embodiments of the invention, examples of which may be illustrated in accompanying figures. These figures are intended to be illustrative, not limiting. Although the invention is generally described in the context of these embodiments, it should be understood that it is not intended to limit the scope of the invention to these particular embodiments.
[0026] Figure 1: Process Flowchart;
[0027] Figure 2: Home Page of Creation Ground (a);
[0028] Figure 3: Home Page of Creation Ground (b);
[0029] Figure 4: Train Your Model Section;
[0030] Figure 5: Train Your Model Section EDA;
[0031] Figure 6: Data Preview;
[0032] Figure7: Statistical Summary & Correlation;
[0033] Figure 8: Feature Distribution;
[0034] Figure 9: Feature Selection;
[0035] Figure 10: Features Datatype Conversion;
[0036] Figure 11: Handling missing values Numerical Columns;
[0037] Figure 12: Handling missing values Categorical Columns;
[0038] Figure 13: Encoding Categorical features;
[0039] Figure 14: Scaling Features;
[0040] Figure 15: PCA & Low Variance Removal;
[0041] Figure 16: Statistical Summary;
[0042] Figure 17: Correlation metrics with Feature Distribution;
[0043] Figure 18: Download Processed;
[0044] Figure 19: Training Model- Algorithm selection, Hyperparameter tunning, Data splitting;
[0045] Figure 20: Classification Report of Validation Data;
[0046] Figure 21: Classification Model – Confusion Matrix; and
[0047] Figure 22: Download Trained Model and Model Structure.
DETAILED DESCRIPTION OF DRAWINGS:
[0048] While the present invention is described herein by way of example using embodiments and illustrative drawings, those skilled in the art will recognize that the invention is not limited to the embodiments of drawing or drawings described and are not intended to represent the scale of the various components. Further, some components that may form a part of the invention may not be illustrated in certain figures, for ease of illustration, and such omissions do not limit the embodiments outlined in any way. It should be understood that the drawings and the detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents, and alternatives falling within the scope of the present invention as defined by the appended claim.
[0049] As used throughout this description, the word "may" is used in a permissive sense (i.e. meaning having the potential to), rather than the mandatory sense, (i.e. meaning must). Further, the words "a" or "an" mean "at least one” and the word “plurality” means “one or more” unless otherwise mentioned. Furthermore, the terminology and phraseology used herein are solely used for descriptive purposes and should not be construed as limiting in scope. Language such as "including," "comprising," "having," "containing," or "involving," and variations thereof, is intended to be broad and encompass the subject matter listed thereafter, equivalents, and additional subject matter not recited, and is not intended to exclude other additives, components, integers, or steps. Likewise, the term "comprising" is considered synonymous with the terms "including" or "containing" for applicable legal purposes. Any discussion of documents, acts, materials, devices, articles, and the like are included in the specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention.
[0050] In this disclosure, whenever a composition or an element or a group of elements is preceded with the transitional phrase “comprising”, it is understood that we also contemplate the same composition, element, or group of elements with transitional phrases “consisting of”, “consisting”, “selected from the group of consisting of, “including”, or “is” preceding the recitation of the composition, element or group of elements and vice versa.
[0051] The present invention is described hereinafter by various embodiments with reference to the accompanying drawing, wherein reference numerals used in the accompanying drawing correspond to the like elements throughout the description. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiment set forth herein. Rather, the embodiment is provided so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those skilled in the art. In the following detailed description, numeric values and ranges are provided for various aspects of the implementations described. These values and ranges are to be treated as examples only and are not intended to limit the scope of the claims. In addition, several materials are identified as suitable for various facets of the implementations. These materials are to be treated as exemplary and are not intended to limit the scope of the invention.
[0052] The present invention provides Creation system for experimentation, creation, and machine learning.
- It offers a user-friendly solution where users with no programming experience can:
- Upload their datasets
- Train machine learning models
- Perform feature selection
- Tune hyperparameters
- Evaluate models
- Download trained models for use
[0053] The prototype is built entirely using Streamlit and popular Python open-source ML libraries like Scikit-learn. Importantly, deployment functionalities (such as API serving or real-time cloud deployment) are NOT part of the current version. The primary novelty lies in simplifying the end-to-end ML pipeline into a clean, no-code interface without sacrificing control over key machine learning steps.
[0054] System Workflow Overview: The system is designed as a linear but flexible ML pipeline consisting of six main steps:
- Data Upload and Preprocessing
- Feature Engineering and Feature Selection
- Model Selection and Training
- Hyperparameter Tuning
- Model Evaluation
- Model Export (Download)
[0055] Each of these steps is automated but provides minimal essential controls to the user for flexibility.
[0056] Data Upload and Preprocessing
[0057] Data Upload:
- The user uploads a .csv or .xlsx file using the Streamlit interface.
- The uploaded data is immediately read into a Pandas DataFrame.
[0058] Automatic Preprocessing:
- Upon data upload, several preprocessing steps are triggered automatically:
[0059] Missing Value Handling:
- Numerical columns: imputed using mean or median.
- Categorical columns: imputed using mode.
- Encoding Categorical Variables: Label Encoding for ordinal data.
[0060] One-Hot Encoding for nominal categorical features.
[0061] Feature Normalization: StandardScaler is used to normalize numerical columns.
- Train-Test Splitting: Data is split into training and testing sets. Stratified sampling is used if the target is categorical.
[0062] Feature Engineering and Feature Selection:
[0063] After preprocessing, feature selection techniques are applied:
[0064] Correlation Analysis:
- Features with correlation above a certain threshold (e.g., 0.9) are eliminated to prevent multicollinearity.
[0065] Principal Component Analysis (PCA):
- Reduces dimensionality by projecting features into a smaller number of components while preserving variance.
[0066] Model Selection and Training
[0067] Based on the problem type (Classification or Regression), users select from pre-configured models:
[0068] Classification Models:
- Decision Tree Classifier
- Random Forest Classifier
- Support Vector Machine (SVM)
- K-Nearest Neighbors (KNN)
- Logistic Regression
[0069] Regression Models:
- Linear Regression
- Ridge Regression
- Gradient Boosting Regressor
[0070] Once the user selects a model, the training happens automatically on the preprocessed and feature-selected dataset.
[0071] Hyperparameter Tuning
[0072] After initial training, the platform automatically optimizes the selected model using either:
[0073] Grid Search: Exhaustive search over a predefined hyperparameter space.
[0074] Random Search: Randomly samples the hyperparameter space. The best parameters are found based on cross-validation performance (e.g., 5-fold CV).
[0075] Example:
- Random Forest: number of estimators, maximum depth
- SVM: C (regularization), kernel type
[0076] Model Evaluation:
- After tuning, the model is evaluated on the test set:
[0077] (a) For Classification Tasks:
- Accuracy
- Precision
- Recall
- F1-Score
- Confusion Matrix
[0078] (b) For Regression Tasks:
- Mean Squared Error (MSE)
- Root Mean Squared Error (RMSE)
- Visual Reports:
- Confusion matrix heatmaps
- Classification report tables
- Regression prediction vs actual plots
[0079] Model Export (Download):
- After evaluation, the user can download the trained model:
- Models are serialized using Pickle (.pkl files).
- Users can use these models later in their own Python scripts for predictions.
[0080] System Architecture:
[0081] The architecture is lightweight and simple:
[0082] Frontend:
- Streamlit (Python-based web framework)
- Machine Learning Engine:
- Scikit-learn for preprocessing, model training, tuning, and evaluation
[0083] Session State:
- Streamlit’s session management used for data transfer between pages.
- Backend Database or Server:
- Not implemented yet.
- Current version uses memory only (session memory).
[0084] Example Walkthrough:
- Imagine a small-business owner wanting to predict customer churn:
- They upload their customer database CSV.
- Missing values are automatically filled.
- Categorical fields like "Gender" and "Membership Type" are automatically encoded.
- Model is selected: Logistic Regression.
- Platform tunes the model hyperparameters.
- User sees the model achieves 88% accuracy.
- Confusion Matrix and classification report appear.
- The user downloads the .pkl model file for future predictions.
[0085] Results from prototype: The dataset used for demonstrating the following outputs is a dummy dataset, intended solely for testing and illustration purposes.
KEY BENEFITS:
[0086] a) No Programming Required:
- Traditional Tools: Most existing tools like Python libraries or cloud ML platforms require users to write code and understand programming logic.
- CreationGround Advantage: Users can perform the entire machine learning process – from loading data to evaluating the model – through a simple, visual interface without writing a single line of code.
[0087] b) Complete Automation of the ML Workflow
- Traditional Tools: Users have to manually handle each stage: cleaning data, selecting features, choosing models, tuning settings, etc.
- CreationGround Advantage: The platform automates all major steps, including:
- Data cleaning
- Feature selection
- Model selection
- Hyperparameter tuning
- Model evaluation
- This saves time and effort, especially for non-experts.
[0088] c) User-Friendly Interface Using Streamlit
- Traditional Tools: Require command-line interfaces or complex notebooks which can be intimidating.
- CreationGround Advantage: Offers a clean, interactive interface built using Streamlit, making it easy to navigate and use even for first-time users.
[0089] d) Suitable for Researchers and Domain Experts
- Traditional Tools: Require collaboration with a data scientist or developer.
- CreationGround Advantage: Researchers from non-technical fields (e.g., agriculture, healthcare, business) can experiment with machine learning models independently to test hypotheses or extract insights from data.
[0090] e) Reduces Human Error
- Traditional Tools: Manual coding increases the chances of mistakes during preprocessing or modeling.
- CreationGround Advantage: Automated workflows ensure consistent and error-free processing.
[0091] f) Open-Source Integration
- Traditional Tools: Some platforms are paid or proprietary.
- CreationGround Advantage: It’s built on top of reliable, open-source tools like Scikit-learn, ensuring cost-efficiency and transparency.
[0092] g) Saves Time and Resources
- Traditional Tools: Require extensive setup and trial-and-error for tuning models.
- CreationGround Advantage: With built-in automation and suggested defaults, users can get usable results much faster.
[0093] h) Supports Multiple ML Models
- Traditional Tools: Users have to manually configure models.
- CreationGround Advantage: The platform offers ready-to-use classification and regression models, along with automatic comparison of results.
[0094] The disclosure has been described with reference to the accompanying embodiments herein and the various features and advantageous details thereof are explained with reference to the non-limiting embodiments in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein.
[0095] The foregoing description of the specific embodiments so fully revealed the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the scope of the embodiments as described herein. , Claims:We Claim:
1) A creation system for experimentation, creation, and machine learning, the system comprising:
- a user interface configured to receive dataset files in standard formats including CSV or Excel;
- a data preprocessing module configured to automatically handle missing values by applying mean, median, or mode imputation, and perform encoding of categorical variables using label encoding or one-hot encoding;
- a feature engineering module adapted to execute correlation analysis and principal component analysis (PCA) to eliminate multicollinearity and reduce data dimensionality;
- a machine learning model selection module configured to select appropriate classification or regression algorithms based on the type of problem;
- a model training module for training selected algorithms on preprocessed and transformed datasets.
2) The system as claimed in claim 1, wherein the model selection module supports classification models including Decision Tree Classifier, Random Forest Classifier, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Logistic Regression, and regression models including Linear Regression, Ridge Regression, and Gradient Boosting Regressor.
3) The system as claimed in claim 1, wherein the system comprises a hyperparameter tuning module configured to optimize selected models using grid search and random search techniques based on cross-validation metrics.
4) The system as claimed in claim 1, wherein the system comprises a model evaluation module configured to calculate metrics including accuracy, precision, recall, F1-score, confusion matrix for classification tasks and mean squared error (MSE), root mean squared error (RMSE) for regression tasks, and visualize results using heatmaps and prediction plots.
5) The system as claimed in claim 1, wherein the system comprises a model export module configured to serialize and export trained machine learning models in Pickle (.pkl) format for external use.
6) The system as claimed in claim 1, wherein the system is implemented using Streamlit for frontend operations and Scikit-learn for machine learning operations including preprocessing, training, evaluation, and tuning.
7) The system as claimed in claim 1, wherein the system enables end-to-end automation of machine learning workflow and is adapted for use by users without any programming or technical knowledge.