A National Level Project Competition on Data Science Challenge 1.0 (DSC 1.0)

Sarvajanik College of Engineering and Technology, Surat in association with IEEE SPS SCET Student Branch Chapter, GS, and IEEE SCET Student Branch organizes, a national-level project competition on ‘Data Science Challenge.’ The goal is to turn data into information, and information into insight. Problem Statements: 1) Forecasting Problem – Time series forecasting can be framed as a supervised learning problem. Similarly, there are many areas like public health systems or crop diseases timely prediction will save the damage or loss in future span. For example, if we want to predict the probability that some event will happen in the future or forecast how many units of a product would be sold over the next six months. Once the model is built and trained using the training data, participants can use it to generate their own probability estimates based on the test data. The test dataset provides similar information in each feature column as that of the training data. Various problems related to forecasting could be: – Forecasting of financial market/stock exchange – Weather forecasting – Covid-19/ disease forecasting for public health system – Earthquake forecasting – Forecasting of Agricultural Crop Disease Risk 2) Biomedical Signal Classification – Biomedical signals are signals primarily used to diagnose or detect specific pathological or physiological conditions. There are different biomedical signals including the electroencephalogram (EEG), the electrocardiogram (ECG), the electromyogram (EMG), the electroneurogram, the electroretinogram, and so on. The proposed problem and, hence, the solution must focus on the classification problem with respect to biomedical signals. An example therein is to predict neurological disorders like epilepsy from EEG signal analysis with competitive accuracy. Classification can be binary like normal or abnormal signal, or it can be a multiclass classification for abnormal signals like for epilepsyictal-interictal-preictal. 3) Optical Character Recognition – Recognizing the characters of any language has been a challenging task in the field of Machine Learning. Optical character recognition using Machine Learning algorithms should solve the problem of recognizing the characters related to Gujarati, Hindi, or English language. The proposed problem must concentrate on the following features: – Use of language data set from the Language Technology Proliferation and Deployment Center or any other authenticated source. – Vowels, part vowels of languages (in the case of Gujarati and Hindi Languages) could be considered for recognition in words. – Handwritten characters could also be used for identification. – Characters combine to form a word and, in turn, a sentence could be identified. – Comparison of results in terms of performance parameters. 4) Digital Image Security – In today’s era of the Internet, Images are a widely accepted form of information communication. However, floating images in their original form across the Internet can lead to potential information breaches. So, the security of images containing critical personal information is of utmost importance. Typically, an image can be secured with any of the mechanisms’ viz. cryptography, steganography, watermarking, reversible watermarking. The proposed problem and, hence, the solution must focus on the security of images while transmitting across the internet. The input for the proposed solution can be any image file containing critical information, and the output is the secured version of the source file which doesn’t reveal that information. Such a secured file can be in terms of other image files or text files. For example, a mammography report of a breast cancer patient should be secured before sending it from the laboratory to a doctor. Rules: – The competition is open for BE/BTech/BCA/B.Sc.(IT and CO)ME/MTech /MCA students. – Participants can form a team of 3 (maximum) and 1 mentor. – Problem statements will be given in two domains, viz. Signal Processing and Image Processing. – All participating teams will send a 3 minutes video clip on idea/work-out related to the given broad problem statement (mentioned below), in the corresponding category by 20th Jan 2022. – Qualified teams will participate in the Semi-final round on 28th Jan, 2022 and will have to work with the exact problem statement provided in the domain of interest. – The final round will be on 29th Jan 2022 (Live Final to be conducted on Google Meet/ On Campus). Winners of the final round will be decided on the basis of performance parameters like accuracy, F1-score, time limit, the computational complexity of the code, use of hardware like Jetson Nano kit, etc. – At least one member of the team should have IEEE SPS membership (preferably, a mentor should be an IEEE SPS member). Else, the participants have to pay INR 500 per team. Guidelines for 3 minutes video: – Work should be unique – The size of the video clip should not exceed 200 MB (SD resolution). – More than 30% plagiarism in the code/information/work may lead to the disqualification of the team with disciplinary actions. – The video clip should clearly explain the work with the associated code and the results – Clearly mention the problem for which the problem is solved and the corresponding dataset used. – A time limit of 3 minutes has to be maintained – Video clip submission format: (.mp4) Agenda: BE/BTech/BCA/B.Sc.(IT and CO)/ME/MTech/MCA students are invited to compete at the national level in a project competition. The competition’s goal was to encourage aspiring technocrats to apply their technical knowledge to develop and demonstrate application-oriented working models in the field of Data Science. Virtual: https://events.vtools.ieee.org/m/299617