Python Based DBMS Assignment
- Country :
Australia
Specific Requirements
Task 1-Image analytic
You are given a sample data set containing images of some popular fruits being sold at Amazing Veges (eg. apple, avocado, banana, nectarine, and pear). The images are stored in images folder. As an Al analyst, you are asked to develop and evaluate a prototype of a fruit recognition system to automatically identify the correct fruit type based on their images using Convolution Neural Network (CNN). You are expected create two CNN models with different architectures (eg, different number of kernels, nodes, layers) to classify the fruit images. You should compare the two models, interpret, summarize the results and recommend a better model for deployment in the intelligence check out system.
Task 2-Text Analytic
You are provided with a large data set containing posts on Twitters from people around the world. The data is provided in the Tweets.csv file You are required to use Python to perform the following analytical and modelling tasks:
- Develop Al pipelines to identify the sentiment of the tweets. You should perform sentiment analysis with two approaches (machine learning based and dictionary based). Compare/Evaluate the performance results with the provided sentiment labels in the data set, discuss the findings, and recommend the best setting for sentiment analysis of the tweets.
- Identify top 10 occurring named entities (eg. person, organization, nationality) in the tweets.
- Perform topic modelling on tweets. Discuss the discovered topics and propose some practical recommendations for WHO from the discovered insights.
Tools:
- Python is to be used for all assignment tasks. You should not modify the data file provided for this assignment before importing it into Python. The preferred platform for this assignment is Google Colab.
The assignment must be prepared using the provided assignment template Lipynb) fileEach of the above tasks should be done using a separate gymb file. Your assignment should contain all necessary codes and ready to run. If you use any new python package, ensure that you include installation code in your jpynb file. All python codes should be ready to execute on Google Collab without any further modification.