Build K-NN and Decision Tree Classifiers for SAT Problem
€30-250 EUR
Plačilo ob prevzemu
Deadline for this project is 13/04/2022.
This project will be coded using Python.
Read the notes section and make sure it is understood.
Using Scikit-learn (and pandas to import the dataset), train and evaluate K-NN and decision tree classifiers using 70% of the
dataset from training and 30% for testing. For this part of the project, we are not interested
in optimising the parameters; we just want to get an idea of the dataset.
Compare the results of both classifiers in a # comment.
Following this, carry out the following methods, and explain the results of each one using a # comment at the start of each code segment:
● Hold-out and cross-validation.
● Hyper-parameter tuning.
● Feature reduction.
● Feature normalisation.
Notes:
Scikit-learn (and pandas just to import the dataset) MUST BE USED, but you can use other libraries along with it.
The dataset you will use is included in the files attached.
The Boolean satisfiability (SAT) problem consists in determining whether a Boolean
formula F is satisfiable or not. F is represented by a pair (X, C), where X is a set of Boolean
variables and C is a set of clauses in Conjunctive Normal Form (CNF). Each clause is a
disjunction of literals (a variable or its negation). This problem is well known as the classic NP-complete
problem.
Recent advances in supervised learning have provided powerful techniques for classifying
problems. In this project, we see the SAT problem as a classification problem. Given a
Boolean formula (represented by a vector of features), we are asked to predict if it is
satisfiable or not.
In this project, we represent SAT problems with a vector of 327 features with general
information about the problem, e.g., number of variables, number of clauses, fraction of
horn clauses in the problem, etc.
The CSV dataset file attached contains 1929 rows and 328 columns. The first 327 columns contain the features. The last column (‘target’) contains the label; 0 for unsatisfiable and 1 for satisfiable.
ID projekta: #33434300
Več o projektu
Dodeljeno:
18 freelancerjev ponuja v povprečju za €144 na tem delu
Hi, How are you? Very happy to bid your project because my skills are fitted in your project. I have 8 years experience in Machine learning ,Deep learning ,NLP and AI. I am very familiar with Generative Adersarial Net Več
Hi there, I can help you complete the Machine learning project. I have read through the projhect description. I will be looking forward to hear from you. Please contact me on PM for details
I can do these project. Python , ML Expert. As 9+ years experiences in these field. I can give good quality work. I have read the guidelines of your work.I believe that i can provide you the best quality works you are Več
Hi In my current Profession -I perfom the DBA and Developer duties in high traffic OLTP environment -Data Mining ,Data Science,Machine Learning -Python programing -data analysis using python and R -Data Scraping u Več
I can implement knn and decision trees for you and provide a report on Jupyter notebook. I am currently free and I can start working immediately. Let's talk over chat so that you assign me the task and I start working. Več
Hi there! Thank you for sharing your project requirement, I have carefully read the project description and I would like to take part in this project. I have the required project skills (Python and Machine Learning (ML Več
Hello,I have read the job description and I am interested in this job. I have 8 years experience in developing products using Machine Learning (ML) and Python. I have read your requirements and am ready to start workin Več
Dear client I have understood your project requirements. I will use decision tree and KNNs algorithm to analyse your data set if you are interested please message me for further details .so I can start working on yo Več