WHAT SHOULD I HACK?

Put your skills to the test and apply Azure AI and Data Services to a solve one or more of the following problem statements. This is an individual hackathon and no teams are required. Projects may use Azure services, open source technologies (including but not limited to frameworks, libraries, and APIs) and physical hardware of your choice. 

To get access to the required tools, Makers must sign up for the Azure free trial or use an existing Azure. New accounts will automatically receive $200 in Azure credits to use towards building submission applications. 

Partner Problem Statements

Accenture

Prize: Hiring Opportunities Mentor: Priyanka Shah - http://www.linkedin.com/in/compuwizpiyu

Problem Statements:

1. Speech to sign language conversion
AI for community initiative. Help the hearing impaired. Participants are expected to train an AI model that can convert text from videos/ presentations to Sign language (using the American Sign language data set)

2. Personalized Gym TrAIner
Get FIT with AI. Build your own AI powered GYM trainer that analyzes your exercise form and reps and tells us if you are really hitting it correctly. Analyze form and count of reps done via exercise classification and pose detection. Open data set available for this. Participants are expected to train a model based on certain exercise postures (yoga, for easier pose detection and availability of dataset).

3. AI SPY : AI powered image caption generation
AI “sees” an image and generates a caption. This is commonly used for auto tagging and generating metadata for images to organize and search images. Also used for marketing campaigns, generating tag lines for products etc. Participants are expected to train a custom vision model trained on available open image sets.

4. Vision COP: image search engine
Build an image search engine for identifying real source of images and searching images based on visual cues instead of text, tags. Content-based image retrieval (CBIR) is a system for retrieving relevant images based on a given image. The system consists of an image query and an image database. This is Heavily used for fake images/news tracking. It is also the basis of recommendations or “images like this… similar images to what you searched”  kind of functionality. Participants are expected to take a test image and show similar images to the test item.

5. AI Professor
Search for documents that talk about the same thing contextually from a set of documents. Participants will be given a data set of some open-source categories (BBC news items). The BBC News Data set can be found here: http://mlg.ucd.ie/files/datasets/bbc-fulltext.zip Participants have to classify the above news articles belonging to certain themes

6. Movies semantic search
Build a knowledge graph based in WIKI music open data set to related to movies. Data set : https://www.kaggle.com/jrobischon/wikipedia-movie-plots The dataset contains descriptions of 34,886 movies from around the world.

Resources available here - https://1drv.ms/w/s!AvJxssZTO60yeU3k3F7U-IsZ7_M?e=YGdK3n 

Just Analytics

Prize: Hiring Opportunities Mentor: Nisha Sullia - https://www.linkedin.com/in/nisha-sullia-3388b49/

Problem Statements:
1. A manufacturing company faced higher complaints in the quality of their product, aircraft parts. The management wants to leverage AI to improve their quality check of products. A smallest dent on the surface can prove to be expensive impacting the integrity of the aircraft. With Image processing, the team wants to identify any defects in the products before approving the quality for shipment. The management would like to understand total defective and non defective parts in a batch of finished product images.

2. Create a power app to allow the car dealership to enter the details of the car sale into the database. The app needs to check for the quality of the data and approve only if the required data meets the data standards else request the dealer to review the data.

Resources available here - https://1drv.ms/w/s!AvJxssZTO60yel4iGPcOAICRlHc?e=EZLvPD 

CIMB 

Prize: Hiring Opportunities Mentor: Chloe Chan Pei Yee, Talent Acquisition, CIMB

Problem Statement:
1. Recruiter receives a huge amount of resumes everyday and the needs of hiring right talent is crucial. How do we use machine learning to screen and rank the top matching profile to expedite our screening process.

Success Criteria
1. Able to shortlist resume based on the skills set needed
2. Able to rank shortlisted resume based one needs of hiring (i.e: years of experience, domain)

You may retrieve the dataset related to this challenge via this link - https://drive.google.com/drive/folders/1W8fO2oqgeWD4iNk1SqYKCXrytzPfnID8?usp=sharing

GBDi (Government Big Data institute) collaborating with Department of Disease Control, Ministry of Public Health of Thailand.

Prize: Hiring Opportunities free data science and/or data engineering training courses by GBDi Mentor: Saranthorn Phusingha, Data Scientist at GBDi https://www.linkedin.com/in/saranthorn-phusingha-8689b4224/

Problem statement:

Thailand has a mission to reduce the number of car accidents and improve road safety. Therefore, it is crucial to understand the causes of the accidentsand identify regions of high risk.

Participants are expected to perform two following tasks:

1. To perform data visualization to analyze trends of accidents, population at risk, risky times or seasons, and possible factors of the accidents. It is highly recommended to perform further analysis to identify other possible important factors, for example, weather, economic deprivation of the regions or alcohol consumption, by using other open reliable datasets (for example, datasets from https://opendata.data.go.th/dataset) to support your analysis.

2. To construct a machine learning model to predict car accidents in each region for each of the next five years. Your analysis should include recommendations on how to prevent car accidents. 

Datasets: https://sipa2016-my.sharepoint.com/:u:/g/personal/saranthorn_ph_depa_or_th/EUVUjK8BvWRKucnUkNHM8OkBmrIS-YQ7KLOSBkNwwqoEWg?e=zeWLEw

HCL Technologies

Prize: Hiring Opportunities Mentor: Mojgan Ghanavati, Lead Data Scientist, HCL Technologies

Problem Statement:
1.Sentiment Analysis
Build a classifier to predict the sentiment of each sentence to positive vs negative sentiments. Three large review datasets from IMDB, Yelp, and Amazon labelled with positive or negative sentiment can be used in this exercise.

2. Credit card defaults 
Participants are expected to train an AI model that can predict the probability of customers default payments in Taiwan. From the perspective of risk management, the result of predictive accuracy of the estimated probability of default will be more valuable than the binary result of classification - credible or not credible clients. The data can be found in this link. 

3. Diabetes is a huge concern area in today’s medical science. Hence, a close observation and prompt preventive / corrective measures are also necessary to have control over this concern. So, a Virtual Diabetes Advisor could be a good interface to take help from.

1. Regular (in batch) update of Diabetic measures for all registered users.
2. System would have defined thresholds depending on age, gender, demographics etc. Thresholds are also subject to revision periodically based on ML based trend analysis of Diabetic measures data.
3. Within 5% range of the threshold, virtual advisor would send proactive voice-based communication to respective users with detailed advices over notification channels like WhatsApp.
4. Users also can reach out to virtual advisor in both text as well as voice mode to seek help / advise / practitioners’ connect etc.
5. The conversations can be audited, and the advisor wisdom can be auto upgraded with time.
6. A Dashboard should be there to represent the healthcare data / conversation data.

Dataset

https://archive.ics.uci.edu/ml/datasets/Diabetes

General Problem Statements:

1. How is Microsoft growing? Help unlock!
"What is a Form 10-K filing?" A 10-K is a comprehensive report filed annually by public companies about their financial performance.Information in the 10-K includes corporate history, financial statements, earnings per share, and any other relevant data. The 10-K is a useful tool for investors to make important decisions about their investments." Can you help decipher important details from the 10-k form for Microsoft?

2. Automating Invoice Reading
Wide World Importers is a logistic company, they get a lot of invoices from different vendors in different formats. They have a team today to manually go through each invoice and scan all invoices and write down important details like Invoice No, supplier details, Item wise details and total amounts. This process sometimes is prone to errors, and also very cumbersome. 

3. Mental health assistant during the pandemic
Covid-19 has put new demands on society. How would you use technology to provide mental health assistance in a virtual scenario where many countries are limiting physical interaction?

Resources available here - https://1drv.ms/x/s!AvJxssZTO60ye74ocwkdIQ2SNnA?e=JMNjtz  

What to Submit:

Provide a URL to your code repository for judging and testing. If your repo is private, it must be shared with elizabete.kalnozola@girlsintech.org before the deadline.

All of the following items must be included in your submission:

1. Submission form 
Submit your project to the Devpost before the deadline and complete the submission form questionnaire. Submissions (including video/code links) are editable until the deadline.
What to include: Problem Statement Partner. Introduction. Purpose & Motivation. How does the application work? How was the application developed? How to use the application? Difficulties & Challenges faced during the design and/or development process? Go-to-Market (How will the application be available to the public, and is it scalable?)

2. Video
Upload a 3-5 min video including the following: Problem Statement Partner. Pitch Deck Presentation. Demo of your product, including an explanation of the solution and function of your application. This video can be submitted as a Youtube link on your Devpost page for your project. Make sure the video is set to public. Note: Any video longer than 5 mins will be automatically disqualified

3. GitHub or other code repo link
Please ensure that this link is publicly shareable so judges can access your project

 4. Provide images, screenshots, & wireframes of your project

***Important: By submitting your project to the Code; Without Barriers Hackathon Devpost, you acknowledge that these materials may be used to promote this hackathon globally.