The Pandemic has impacted education - classes have moved online, students have been isolated on screens and coping with this change. Despite the challenges, the digital school has the potential to transform education. How can we empower students and teachers in this new digital school paradigm. In this challenge, we are inviting AI-powered solutions for the digital school of tomorrow.
DATASET: Feel free to use any dataset of your choice.There is no restriction and you can use any data set. Please see the section - DATASETS below for sample datasets to help as a reference. Here are sample areas you could choose to tackle in this challenge. Feel free to come up with your own ideas as well. 1. Higher Education and Career Recommendation 2. Mental Health Monitor and Virtual Companion 3. Adaptive Learning Curriculum 4. Class availability scheduling for social distancing 5. Compliance of COVID guidelines - masking, distancing, temperature Please see below for guidelines and reusable libraries to jumpstart your solution. This kit provides reference to open-source libraries which can be reused as core building blocks for creating a predictive solution. You may use any other open-source libraries also as relevant to your solution. Reusability is a key design principle and will be scored positively in your submission. These reference reusable libraries are spread over functions in Data Analysis and Mining, Data Visualization, Machine Learning, and other key areas to build AI solution. Below are the guidelines for creating your submission kit for this challenge. 1. See Product Tour > Creating a kit from the kandi header. This will guide you on creating your kit. 2. Your submission kit should contain the kit heading/ name, description of the solution, and other relevant information. 3. Create groups with logical names and add the libraries to the respective sections. 4. You solution can be built with any libraries other than the libraries provided here for reference. 5. The project source library for the solution built in the hackathon should be hosted in GitHub and listed in your kit under 'Kit Solution Source' section. 6. Any deployment instructions should be added under 'Kit Deployment Instructions' section of the kit. 7. Add any additional information, links under the kit description or group descriptions.
https://data.ed.gov/ https://data.world/datasets/education https://data.gov.in/sector/higher-education https://github.com/mdsaifk/Student-Dropout-Prediction/tree/main/Data https://github.com/hilmarh/student-dropout-prediction/tree/master/datasets https://github.com/iampratheesh/Student-Dropout-Prediction/blob/master/student%20info.csv https://www.kaggle.com/spscientist/students-performance-in-exams https://www.kaggle.com/aljarah/xAPI-Edu-Data https://www.kaggle.com/janiobachmann/math-students?select=student-mat.csv https://www.kaggle.com/kwadwoofosu/predict-test-scores-of-students https://www.kaggle.com/namanmanchanda/entrepreneurial-competency-in-university-students https://www.kaggle.com/uciml/student-alcohol-consumption?select=student-por.csv https://www.kaggle.com/passnyc/data-science-for-good https://www.kaggle.com/landlord/education-and-covid19
VSCode and Jupyter Notebook are used for development and debugging. Jupyter Notebook is a web based interactive environment often used for experiments, whereas VSCode is used to get a typical experience of IDE for developers.
Jupyter Interactive Notebook
Jupyter Notebook 9914 Version:v7.0.0a18 License: Permissive (BSD-3-Clause)
Data Analysis and Mining
Data Mining and Analysis plays vital role in Predictive Analytics. It lets you inspect, cleanse, explore, manipulate, transform your data to identify hidden patterns and relationships in data. You can make use of these popular libraries to model the solution.
Flexible and powerful data analysis / manipulation library for Python, providing labeled data structures similar to R data.frame objects, statistical functions, and much more
Python 37439 Version:v2.0.0rc1 License: Permissive (BSD-3-Clause)
The fundamental package for scientific computing with Python.
Python 23036 Version:v1.24.2 License: Permissive (BSD-3-Clause)
Data Visualization helps you depict insight found in data. Avail the libraries added here to represent identified patterns and relationships from data graphically for better understanding and presentation.
matplotlib: plotting with Python
Python 17111 Version:v3.7.1 License: No License
Statistical data visualization in Python
Python 10513 Version:v0.12.2 License: Permissive (BSD-3-Clause)
The open and composable observability and data visualization platform. Visualize metrics, logs, and traces from multiple sources like Prometheus, Loki, Elasticsearch, InfluxDB, Postgres and many more.
TypeScript 54661 Version:v8.5.22 License: Strong Copyleft (AGPL-3.0)
Libraries in this group are used for analysis and processing of unstructured natural language. The data, as in its original form aren't used as it has to go through processing pipeline to become suitable for applying machine learning techniques and algorithms.
💫 Industrial-strength Natural Language Processing (NLP) in Python
Python 25655 Version:v3.5.1 License: Permissive (MIT)
GloVe model for distributed word representation
C 6255 Version:1.2 License: Permissive (Apache-2.0)
Stanford CoreNLP: A Java suite of core NLP tools.
Java 8907 Version:v4.5.4 License: Strong Copyleft (GPL-3.0)
Simple, Pythonic, text processing--Sentiment analysis, part-of-speech tagging, noun phrase extraction, translation, and more.
Python 8472 Version:0.7.0 License: Permissive (MIT)
Multilingual Sentence & Image Embeddings with BERT
Python 9823 Version:v2.2.2 License: Permissive (Apache-2.0)
🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
Python 88853 Version:v4.27.4 License: Permissive (Apache-2.0)
Image Analysis plays vital role in Visual Analytics. It lets us inspect, cleanse, explore, augment and transform images to prepare data for training and prediction.
Image processing in Python
Python 5321 Version:v0.20.0 License: Others (Non-SPDX)
The Open Source Framework for Machine Vision
Python 2590 Version:Current License: Permissive (BSD-3-Clause)
Machine learning algorithms and techniques
To build a model for Predictive Analytics, you can apply traditional machine learning algorithms and techniques using the most popular scikit-learn. Or you can build your own neural network to implement deep learning techniques by using the library of your choice from this section.
scikit-learn: machine learning in Python
Python 53572 Version:1.2.2 License: Permissive (BSD-3-Clause)
An Open Source Machine Learning Framework for Everyone
C++ 172599 Version:v2.12.0 License: Permissive (Apache-2.0)
Tensors and Dynamic neural networks in Python with strong GPU acceleration
C++ 64612 Version:v2.0.0 License: Others (Non-SPDX)
A library for transfer learning by reusing parts of TensorFlow models.
Python 3284 Version:v0.13.0 License: Permissive (Apache-2.0)
Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit
C++ 17339 Version:v2.7 License: Others (Non-SPDX)
Theano was a Python library that allows you to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays efficiently. It is being continued as aesara: www.github.com/pymc-devs/aesara
Python 9691 Version:Current License: Others (Non-SPDX)
Request servicing via REST API
Web frameworks help build serving solution as REST APIs. The resources involved for servicing request can be handled by containerising and hosting on hyperscalers.
FastAPI framework, high performance, easy to learn, fast to code, ready for production
Python 56062 Version:0.95.0 License: Permissive (MIT)
The Python micro framework for building web applications.
Python 62385 Version:2.2.3 License: Permissive (BSD-3-Clause)