Just a few days away now, Machine Learning Developers Summit, which is to be held on 22-23 Jan in Bengaluru and on 30-31 Jan in Hyderabad, has created a buzz around the tech community. With MLDS, Analytics India Magazine aims to bring in researchers and innovators together on one platform, where they will be presenting their research papers on various topics like machine learning, deep learning, and robotic process automation (RPA).
In this article, we list down the top paper presentation at MLDS that the attendees should not miss:
1| A Novel Approach For Product Recommendation Engine Using Graph Database
Presenter: Naman Mishra, Data Scientist at Genpact
About: In this paper, the author will be presenting a novel approach using graph algorithms for building a product recommendation solution for a publishing company. He will be talking about the developed approach that focuses on the popular books and courses inside a local community identified by the graph algorithms to generate recommendations.
2| Emotional Stress Detection Using Deep Learning
Presenter: Nithya Vasudevan Analyst, Data Science at Verizon
About: In this paper, the author will be presenting an idea where they use neural network architectures for attention mechanisms to spot the people who are suffering from prolonged stress. He will also talk about the offered solution that can be used for the long term sufferers by predicting and analysing their emotions using brainwaves recorded through Neurosky Brainwave Headset.
3| Revolutionalising Safety In Railways Using Computer Vision
Presenter: Vibhav Patil, Senior Machine Learning Engineer at Racetrack.ai
About: In this paper, the author will be presenting a complete solution of critical issues like rail accident, dying of animals on the railway tracks, etc. using computer vision technology. He will also be talking on methods based on spatial-temporal relationships, long-range cameras as well as to object detection using deep learning models.
4| A Heuristic Optimisation Solution For The Selection Of The Transformation Functions For Media Channels In Marketing Mix Modelling (MMM)
Presenter: Madhav Kaushik, AVP at Analyttica Datalab
About: In this paper, the author will be presenting a heuristic optimisation methodology for the selection of the transformation functions for the media channels. He will also be explaining how it is automated and how it has been developed into a prototype that can be applied in similar situations towards the process of measuring “Above the Line” (ATL) marketing effectiveness and optimisation.
5| NLP Driven Qualitative Assessment Of Product Description
Presenter: Siddharth Vij, Data Scientist at Happiest Minds
About: In this paper, the author will be presenting an automated mechanism using lexical, syntactic, and contextual NLP techniques to assess the quality of product descriptions by scoring the usage of personal, sensorial, functional and superlatives in a description.
6| Artificial Intelligence For Simplified Deployments
Presenter: Suchit Mathur, Product Expert at SAP Labs India
About: In this paper, the author will be presenting an approach to automate custom manual actions using RPA framework and machine learning that enabled natural language processing system for providing inputs to a system using voice, text, etc.
7| Measuring Digital Marketing Effectiveness Using Incrementality
Presenter: Shubham Gupta, Data Scientist at MiQ Digital
About: In this paper, the author will be presenting different approaches to calculate incremental lift that can be implemented in the digital marketing ecosystem such as viewability. He will also talk about the concepts of test environment setup, randomisation, bias handling, hypothesis testing, primary output, and understanding different ways of using the output of incremental lift — for instance, strategy planning and optimisations, and achieving higher campaign efficiency, among others.
8| Automated Short Answer Grading Using Conventional And Modern NLP Techniques
Presenter: Bharath Kumar Bolla Senior Data Scientist at Happiest Minds
About: In this paper, the author will be presenting research into Automatic Short Answer Grading (ASAG) by combining conventional and advanced Natural Language Processing (NLP) methods which include various embeddings ranging from word level, sentence level and contextual. He will also be presenting a comprehensive evaluation of various embedding techniques (word2vec, FastText, ELMo, Skip-Thoughts, Quick-Thoughts, FLAIR embeddings, InferSent, Google’s Universal Sentence Encoder and BERT) with respect to short text similarity.
9| Building An AI-driven Logistics Platform
Presenter: Rishit Jain, Product Manager, Data Science and AI Strategy at Delhivery
About: In this paper, the author will be presenting key tenets and levers of building an AI platform from scratch and specific challenges that are unique to AI products. He will be talking on factors of system driven allocation such as ensuring fairness and equal opportunity, utilising local ground intelligence, and complexity of delivering a shipment.
10| Predicting Product Success Using AI In Video And Audio Analytics
Presenter: Govind Maheswaran, Senior Consultant at Ernst & Young
About: In this paper, the author will be presenting a methodology of automating the role of a moderator in the Focus Group discussions using artificial intelligence, where the proposed solution uses machine learning and deep learning to process the video and audio streams of a Focus Group.
11| Leveraging BERT + Deep Learning For Impactful Analysis On Streaming News And Events
Presenter: Akshay Sharma, Sr. Data Scientist at Tidyquant
About: In this paper, the author will be presenting a model which is capable of processing huge amounts of data from multiple sources like tweets, websites, documents, etc., and provides deep insights by identifying the context of the data which has been provided to it. He will also be talking about how the solution works over Bidirectional Encoder Representations (BERT), the neural network-based technique for Natural Language Processing (NLP) pre-training.
12| Partitioning Nearest Neighbour Approach To Regression Variation Improvement In Tree-Based Approaches
Presenter: Abhinav Mathur, Data Scientist at Clinton Health Access Initiative
About: In this paper, the author will be presenting a hybrid approach of using two intuitive and explainable algorithms, CART 2 and k-NN 3 regression to improve the generalisations and sometimes the runtime for regression-based problems.
13| Algorithm To Recommend Corrective Actions For Yaw Misalignment In A Wind Turbine
Presenter: Malavika Peedinti, Assistant Manager – Data Analytics at BLP Clean Energy
About: In this paper, the author will be explaining how machine learning is being used to detect yaw error right after its onset, identifying the root cause and quantifying the consequent energy loss. Peedinti will also discuss the detection of yaw misalignment is done by analysing the wind direction, nacelle position, and yaw angle data captured by the Supervisory Control and Data Acquisition (SCADA) system installed at the wind power plant.
14| Detecting and Predicting Price Change Acclimation
Presenter: Kajal Anajwala, Senior Manager, Global Advanced Analytics – Diageo
About: In this paper, the author will be presenting a model which will detect the time taken for demand to come back to a normal level after the price change. It then predicts how long it takes for demand to come back to a normal level if there is a plan for a price increase in the future.
15| Approaches For Predicting Potential Cancerous Cell Formation in the Fetus
Presenter: Debashish Banerjee, CEO at Blackstone Synergy
About: In this paper, the author will be presenting mathematical models exploring the parametric influences of the pathological changes, the hormonal changes, the relative coordinates of the fetus, the relative growth ratio of the brain and the body, and finally the functional MRI derivatives for cell morphology and cytoplasmic changes that signal potential mutants in the embryonic development evolution.
16| Content And Author Identification In Indian English, Hindi and Bangla
Presenter: Subhabrata Banerjee, Computational Linguist at HCL Technologies.
About: In this paper, the author will be presenting a combined solution in Indian English, Hindi and Bangla for content detection and author identification. Banerjee will also be discussing the work exploits three online monolingual corpora of plain text, as well as Named Entity, annotated text.
17| Machine Learning Application To Create Experimental Learning Models For Personal Finance
Presenter: Bharat Shah, Chartered Accountant and ISB AMPBA alumnus
About: In this paper, the author will be presenting a predictive model which applies Unsupervised Learning and Supervised Learning to achieve different clusters of income and expenses to help in identifying the pattern of income and expense. He will also talk about how it provides the ability to visualise someone’s income and expense pattern for 5-10 year.
18| Smart Job Order Prioritization with AI
Presenter: Shiva Tyagi, Machine Learning Engineer at TCS
About: In this paper, the author will be presenting a solution which focuses on helping recruiters prioritise job requests based on probability score of request completion thereby increasing fill rate as the number of job request completed and reduce the time-taken, ensuring higher value return with relatively less effort.
19| Domain-Specific Word Segmentation and Hierarchy Detection using NLP Algorithm
Presenter: Prakash Selvakumar AVP at Genpact
About: In this paper, the author will be presenting a machine learning approach to build a word segmentation algorithm and also find the hierarchical structure in the text (for example Header1 or Header2, etc.) The paper focuses on a machine learning approach to perform word segmentation and Hierarchy detection on medical documents (in the form of editable digital materials like PDFs).