Name: Ankit Mishra
Profile: Associate SDE
Company: AFour technologies
Email: mishra.ankit2232@gmail.com
Mobile: +91-7004052008
Skill
PythonAbout me
An accomplished Associate Software Development Engineer specializing in AI/ML, NLP, and Generative AI, with a Master's in Computer Applications from Goa University. Possessing over 1 year of experience in backend development and proficient in Python, AWS, and various development tools. Published researcher with notable contributions to sentiment analysis and hope speech detection. Recognized for winning competitions like Afourathon 4.0 and securing top ranks in HASOC 2020 and EACL 2021. Seeking impactful roles in cutting-edge AI projects to drive innovation and deliver meaningful results.
Portfolio
I have extensively explored and implemented a wide array of traditional machine learning techniques. Currently, my focus lies on staying abreast of the latest advancements in Generative AI approaches, particularly in the realm of Large Language Models (LLMs) and Small Language Models (SLMs). My aim is to continuously update my skill set to leverage cutting-edge techniques and methodologies in pursuit of solving complex problems effectively and efficiently.
Here are some of the projects done by me.
PII Data Discovery (Asset Project)
AI SuperBot (AFourathon 4.0)
Enviate (Client)
iSYnopsis(Asset Project)
Certificates
Internship Certificate
Afourathon 4.0
Machine Learning Zero to GBMs
Swaksh Bharat Student Internship (2021)
ELECTHON
Research
I have Participated in some data science competitions and participated in Globally recognised conferences Like ACL and FIRE.
Natural Language Processing
Identifying offensive content in Indo-European languages
Human behaviour remains the same whether it is a physical or cyber world. They express their emotions like happy, sad, angry, frustrated, bullying, and so on at both places. To express these emotions in cyberspace one of the way is a text post. The impact of these posts lasts forever on social media sites like Twitter, Facebook, and so on. Some posts that contain hate and offensive content affect victims badly and drag them into mental illness. The current paper aims to identify such hate and offensive posts using deep learning-based models such as CNN, and LSTMs. The Twitter posts in English, Hindi, and German languages used in this study are a part of HASOC-2020 competition. The model submitted for English sub-task A outperformed all other models submitted in the competitions by securing the 1st rank and F1-macro average score of 0.5152.
Computer Vision
Identifying Troll Meme in Tamil using a hybrid deep learning approach
Social media are an open forum that allows people to share their knowledge, abilities, talents, ideas, or expressions. Simultaneously, it also allows people to post disrespectful, trolling, defamation, or negative content targeting users or the community based on their gender, race, religious beliefs, etc. Such posts are available in the form of text, image, video, and meme. Among them, memes are currently widely used to disseminate offensive material amongst people. It is primarily in the form of pictures and text. In the present paper, troll memes are identified, which is necessary to create a healthy society. To do so, a hybrid deep learning model combining convolutional neural networks and bidirectional long short term memory is proposed to identify trolled memes. The dataset used in the study is a part of the competition EACL 2021: Troll Meme classification in Tamil. The proposed model obtained 10th rank in the competition and reported a precision of 0.52, recall 0.59, and weighted F10. 3.
Natural Language Processing
hope speech detection in YouTube multilingual comments
Language as a significant part of communication should be inclusive of equality and diversity. The internet user’s language has a huge influence on peer users all over the world. People express their views through language on virtual platforms like Facebook, Twitter, YouTube etc. People admire the success of others, pray for their well-being, and encourage on their failure. Such inspirational comments are hope speech comments. At the same time, a group of users promotes discrimination based on gender, racial, sexual orientation, persons with disability, and other minorities. The current paper aims to identify hope speech comments which are very important to move on in life. Various machine learning and deep learning based models (such as support vector machine, logistics regression, convolutional neural network, recurrent neural network) are employed to identify the hope speech in the given YouTube comments. The YouTube comments are available in English, Tamil and Malayalam languages and are part of the task “EACL-2021: Hope Speech Detection for Equality, Diversity and Inclusion”