A facial analysis system that predicts personality traits and characteristics from facial images using computer vision and machine learning techniques.

In recent days, Social Media usage has been increased drastically. Especially, Instagram is one of the top platforms for visual content sharing. In this study, we are inferring the behaviour for the Big Five personality traits through the Instagram data. For this analysis, we are using the Shared Images, Content of the posts and Key Performance Indicators for each post. As part of Image analysis, we take the HSV (Hue, Saturation, and Value) color space for images and deduce the relationship between the characteristics of the users and color patterns. Usually, people will adjust the appearance of the images and then post it. So this assessment helps to understand the type of filters & colors is opting by each personality. From the Linguistic information, we intended to gather the significant unigram/bigram words for each personality and trace influence over the other characteristics. The investigation is ap-proached by implementing the different categorical algorithm and validating the model accuracy. Fi-nally, with the help of metrics, we could understand the activeness & social interaction for each per-sonality traits. The regression model is used to extract Features from the quantitative data. Thus, this paper will give the depth of information about actions & pattern of thinking of each personality and correlation between the Instagram user’s data and characteristics. This information could be used in different domains. Commonly, this kind of analytics helps to target the correct users when setting business campaigns.

Tool for reading NZSL-Dictionary dataset, and using PoseNet ML model to extract information and images from video of NZSL sign performances, to generate datasets to train CNNs to recognise traits of visual signed languages

Biometric system is a pattern recognition system that works by collecting biometric data from a user, extracting a feature set from that data, and comparing that feature set to a database template set. Through this paper, we propose a biometric recognition system based on iris recognition. Iris is the most secured and unique biometric trait among other biometric traits. In our work, we have proposed a modified Hough Transform and considered the Mini-VGG Net model without its weights, and trained the network to obtain the best features. Using the Neural Networks, we have performed the classification and obtained Accuracy, Precision, and Recall of 98%, 0.99, and 0.99 respectively. Our experiments were performed on the CASIA Version-1, which includes 756 samples of iris in 108 folders with 7 samples having dimensions 280X320 each.

Our framework (1) quantify the relative influences of different low level features on the interview outcome, (2) learns regression models to predict interview ratings and the likelihood of hiring using automatically extracted features, and (3) predicts several other high-level personality traits such as engagement, friendliness, and excitement. One of our objectives is to extend the existing automated conversation systems by providing feedback on the overall interview performance and additional high-level personality traits.

An AI-powered LinkedIn personality analyzer that extracts insights from public profile data using NLP and large language models to generate behavioral, communication, and professional trait analysis.

LFM-1B user subset with extracted style- and topic features for user trait prediction (+weka models)

Skillbrew.AI processes live webcam video to extract 10 facial features and scores 8 behavioral traits in real-time. Each trait has heuristic weights validated against the FER2013 emotion dataset, blended with a Random Forest model (HYBRID mode after training).

Research implementation demonstrating cross-architecture persona vector transfer for controlling personality traits in large language models. Extracts vectors from Qwen2.5-7B and applies them to steer GPT-OSS 20B behavior.

This app will analyzes input handwriting image using image processing techniques, extract features and based on them it will output the most possible component among the five components of Five-Factor model personality taxonomy. In addition to that it will provide a fully detailed report about writer's personality traits based on the extracted handwriting features and provide solutions to get rid of negative personality traits.

Detecting malicious Twitter bots using machine learning involves collecting data on Twitter accounts, extracting features like tweet content, account metadata, and behavioral patterns, and training models to identify bots. Features include linguistic traits, network analysis, and posting behavior.

This project analyzes online customer purchasing behavior to extract actionable insights for marketing, personalization, and revenue optimization. Using data cleaning, visualization, and predictive modeling, it examines how user traits, session activity, and product interactions influence purchase probability and spending.

An end-to-end Twitter analytics project using NLP to extract sentiment, personality traits, and engagement insights. The project applies Big Five personality profiling with Hugging Face models, processes tweets using Python and Tweepy, and presents insights through an interactive Streamlit dashboard.

This project presents a machine learning and deep learning classifier for handwriting analysis, focusing on the letter "t" as a personality indicator. Using CNN models, machine learning, and OpenCV, the system extracts features like size, slant, pressure, and spacing to classify 19 traits, achieving over 95% accuracy.

The heading date and effective tiller percentage are important traits in rice, and they directly affect plant architecture and yield. Both traits are related to the ratio of the panicle number to the maximum tiller number, or simply called panicle ratio (PR). In this study, an automatic PR estimation model (PRNet) based on a deep convolutional neural network was developed. Ultra-high-definition unmanned aerial vehicle (UAV) images were collected from cultivated rice varieties planted in 2384 experimental plots in 2019 and 2020 and in a large field in 2021. The determination coefficient between estimated PR and ground measured PR reached 0.935, and the root mean square error (RMSE) values for the estimations of the heading date and effective tiller percentage were 0.687 days and 4.84%, respectively. Based on the result analysis, various factors affecting PR estimation and strategies for improving PR estimation accuracy were investigated. The satisfactory results obtained in this study demonstrate the feasibility of using UAV and deep learning techniques to replace ground-based manual methods to accurately extract phenotypic information of crop micro targets (such as grains per panicle, panicle flowering, etc.) for rice and potentially for other cereal crops with future research.

Attending meeting-filled days wearing blue-light glasses, with a pizza slice in one hand and a dumbbell in the other, striking a perfect balance of endorphins and staying healthy, you wonder how the best developer ever gets any work done. The secret is, he knows coding is only the last-mile to any solution. We all know one engineer who seems to have the answers to all and everyone’s questions — or, more importantly, knows how to dig out answers. The one who knows exactly how to deal with that difficult issue you’re facing, or, can elegantly model a complex domain and implement logic in a readable, understandable, and maintainable manner. Difficult as it may be, I’ll attempt to break down the traits of the best programmer ever. Business-first mindset. Knowing the business is any developers’ first job. The best developer produces solutions and writes code that he knows will positively benefit the business. Knowing the what and why leads to effortless implementations (the how). It becomes easier to write great code when you know which internal code qualities are appreciated. The best developer doesn’t strive to produce the most performant code if the performance for a specific feature is not prioritized or necessary. The “everything’s an object“ mindset. The best developer obviously has an object-oriented approach and follows the mantra “everything’s an object.” He uses domain models to capture the complexity by creating conceptual blueprints of the world viewed thru the lens of their domains’ context. Coding is the last activity and is only carried out after having a thorough understanding of the domain, its events, and how different parts interact. Common issues are dealt with by tailoring known design patterns, allowing them to be implemented effortlessly. “Everything’s an object” is the perfect mindset that facilitates creating understandable and maintainable applications with ease. Each else-if and switch case is extracted into its own class. Implicit dependencies are levitated to explicit, and interfaces are preferred to concrete classes. The chase for flexible code is never-ending. Relentlessly accepts new tasks. The only tasks the best developer dreads are those poorly understood. Those that are poorly understood need clarification immediately. Assuming business requirements is always a terrible idea. The best developer is not shy of setting up meetings with business people. Despite being conscious of the shame, it entails as fellow engineers have read the latest and greats blog post spurting out nonsense like “developers should not attend meetings.” Project-wide refactorings are a breeze: extract a class there, implement an interface here, turn chaos into order with a set of nifty design patterns. Every refactoring activity leaves the codebase in a cleaner state. Commenting with style and purpose. The best developer doesn’t buy into the fairytale that code must be self-documenting. Code is rarely self-documenting. It needs comments. Documentation is to a large extend about “why” a piece of code exists, and there’s only so-much context a perfectly named class, method, or variable can capture. He doesn’t comment on language semantics, syntax, or even if a ternary turns crazy. He respects his peers enough to assume they know how to read code. On the other hand, commenting on the reasoning and intent is done so thoughtfully. Everyone does not have the same thought patterns and domain experience. Both are required to understand why some code exists. This perfectly commented method below exemplifies my point. Perfectly commented method in the aspnet core source code. Perfectly commented method in the aspnet core source code. It takes time to write good code doc comments. But it takes even longer for the uninitiated to understand why some code exists. Nicklas Millard is a software development engineer in one of the fastest-growing banks, building mission-critical financial services infrastructure. Previously, he was a Big4 Senior Tech Consultant developing software for commercial clients and government institutions.

A minimal toolkit for extracting and analyzing trait vectors from transformer language models.

AI-based pipeline for extracting standardized traits from species descriptions using language models. This tool encodes binary, numeric, and categorical traits into structured trait matrices, enabling users to generate dichotomous keys and perform ecological and evolutionary analyses of diversity, trait distribution, and uniqueness across species.

Utilized Facebook API to extract user’s posts to predict personality by SVM model based on big 5 personality traits (OCEAN model of Psychology).

A personalised Python application that predicts personality traits based on the Big Five personality model while extracting and analyzing resume data.

This project leverages Natural Language Processing (NLP) and Machine Learning (ML) techniques to analyze resumes (CVs) and predict an individual's personality traits. The model extracts key features from resumes, applies text processing techniques, and classifies personality traits based on predefined psychological frameworks

A toolkit for controlling personality traits in language models through activation steering. This repository contains code for extracting, validating, and manipulating personality vectors in Gemma-2-2B-IT using the Big Five personality framework.

A machine learning based Personality Prediction System that analyzes resumes (CVs) using NLP techniques to predict personality traits. The system extracts text features from resumes, processes them using Python libraries, and predicts personality types through trained ML models.

Personality Prediction Using Handwriting & LLMs is an AI-powered system that analyzes handwriting samples to predict personality traits based on the Big Five model (Agreeableness, Conscientiousness, Extrav The workflow combines OCR (to detect and extract handwriting), a deep learning model (EfficientNet-B0 for classification), ersion, Neuroticism,

Personality Prediction Using Handwriting & LLMs is an AI-powered system that analyzes handwriting samples to predict personality traits based on the Big Five model (Agreeableness, Conscientiousness, Extrav The workflow combines OCR (to detect and extract handwriting), a deep learning model (EfficientNet-B0 for classification), ersion, Neuroticism,

tool that generates a detailed persona profile for any Reddit user by analyzing their public comments and submissions. It leverages advanced language models and semantic search to extract demographic and psychographic traits, providing direct quotes and source links for each insight.