Naive bayes classifier jobs
I’m building a Python solution that automatically flags spam in my personal inbox. The pipeline should read raw message text, run it through classic NLP cleaning—then let a Naive Bayes model trained on TF-IDF features decide what is junk and what I actually want to see. I will be working exclusively with publicly available email corpora, so nothing proprietary needs to be handled. Key pre-processing I need implemented: • Tokenization • Stopwords removal • Stemming or lemmatization The stack is already chosen: Scikit-learn, Pandas and NumPy under Python. Once training is complete, I’d like a concise script or notebook that can: 1. Load a fresh batch of .eml or plaintext messages. 2. Output a CSV (or similar) labeling each as spam ...
I’m building a web feature that can read whatever a visitor types into the site’s chat box, gauge their emotional tone through text-analysis NLP, and immediately queue up music that matches or lifts that mood. Here’s the flow I have in mind. The user interacts with the live-chat on our site. Every message is piped to your classifier, which returns a mood label or valence/arousal scores. Based on that output, the system picks an appropriate track or playlist from a streaming service (Spotify or a royalty-free catalogue—whichever is easier to wire up first) and starts playback without noticeable delay. I’m set on text analysis as the detection method; no facial or voice inputs are needed right now. Likewise, I don’t need social-media or email min...
...Sana. I’m launching an Indian-based brand that celebrates youth, culture, freedom of expression and raw creativity. I need is a logo that captures that spirit in a Bold and vibrant style. I’m picturing a lively mix of bright, energetic hues paired with a memorable symbol and text working together rather than separately, so the name and the icon feel inseparable. I have been quite fascinated by Naive design and maximalism. Need assistance in understanding what would be best for the brand. I will be using the logo across merchandise, social channels and large-format prints, so scalability in Adobe Illustrator or an equivalent vector tool is a must. Please keep colours in RGB and CMYK palettes ready for both digital and print use. Deliverables • Primary logo (ic...
...detection or enhancement features are required. You will start from an existing dataset of labelled product images (I’ll supply a download link once we begin). A lightweight, production-ready model built in Python with either TensorFlow or PyTorch is preferred; transfer-learning from a well-known backbone (e.g., ResNet, EfficientNet, MobileNet) is perfectly acceptable so long as the final classifier is accurate and quick to infer on standard CPU hardware. Key deliverables: • Clean, commented source code for training and inference • The trained model weights (or exported SavedModel / .pt file) • A short README explaining environment setup, how to retrain with new data, and a one-line CLI or REST example for running predictions • Basic metrics re...
...testing. What You Will Do: Design a specific structural engineering problem (e.g., optimizing a truss under specific load envelopes, calculating Euler buckling limits, or solving for indeterminate structures). Provide the "Oracle Solution": the step-by-step, 100% correct mathematical derivation and the final numerical answers. Introduce realistic constraints or edge cases that would trick a naive automated system (e.g., ensuring a system fails due to buckling before it yields to stress). Required Skills & Qualifications: Advanced Structural Analysis: Deep understanding of load distribution, structural elements, and internal force calculations. Indeterminate Structures: Ability to calculate statically indeterminate structures where basic equilibrium equations ...
...(such as temperature, pressure, humidity, and location-based data) and outputs wind speed/direction predictions, useful for applications like weather forecasting, renewable energy planning (wind turbine placement), and agricultural planning. Skills demonstrated: Data preprocessing, regression/classification modeling, feature engineering, Flask web integration, model deployment. Student Admission Classifier Developed a machine learning classification model to predict student admission outcomes based on academic and personal parameters (such as GPA, test scores, extracurricular activities, and other relevant features). Deployed as a web application where users input their details and receive a predicted admission likelihood/outcome, useful for ed-tech platforms, counseling servic...
Need Freelance GK question Content Classifier Work: Categorize approximately 10,000 GK MCQs into predefined subjects and topics. Questions are one-liner objective questions. No content creation required. Preferred: UPSC/WBCS/SSC aspirants Experience in educational content writing in govt exam preparation(Optional) Selection Process: Paid sample test of 200 questions. Accuracy and consistency will be evaluated before assigning the full project. Deliverable: Work Involves: • Reading and understanding each MCQ. • Classifying questions under predefined Subjects and Topics. • Organizing questions chapter-wise for book publication. • Removing duplicate and near-duplicate questions. • Ensuring that questions are placed under the most appropriate topic. &bu...
Need Freelance GK question Content Classifier Work: Categorize approximately 10,000 GK MCQs into predefined subjects and topics. Questions are one-liner objective questions. No content creation required. Preferred: UPSC/WBCS/SSC aspirants Experience in educational content writing in govt exam preparation(Optional) Selection Process: Paid sample test of 200 questions. Accuracy and consistency will be evaluated before assigning the full project. Deliverable: Work Involves: • Reading and understanding each MCQ. • Classifying questions under predefined Subjects and Topics. • Organizing questions chapter-wise for book publication. • Removing duplicate and near-duplicate questions. • Ensuring that questions are placed under the most appropriate topic. &bu...
Need Freelance GK question Content Classifier Work: Categorize approximately 10,000 GK MCQs into predefined subjects and topics. Questions are one-liner objective questions. No content creation required. Preferred: UPSC/WBCS/SSC aspirants Experience in educational content writing in govt exam preparation(Optional) Selection Process: Paid sample test of 200 questions. Accuracy and consistency will be evaluated before assigning the full project. Deliverable: Work Involves: • Reading and understanding each MCQ. • Classifying questions under predefined Subjects and Topics. • Organizing questions chapter-wise for book publication. • Removing duplicate and near-duplicate questions. • Ensuring that questions are placed under the most appropriate topic. &bu...
...machine-learning model that can automatically flag fraudulent activity. The model must correctly recognise the three problem categories—Phishing, Robocalls and Telemarketing scams—without human intervention. What I expect you to handle: • Pre-processing: clean the audio and extract features (e.g., MFCCs or spectrograms) that capture speaker and content cues. • Modelling: design, train and fine-tune a classifier; CNN, RNN, Transformer or a hybrid approach is acceptable if it improves accuracy. • Evaluation: deliver precision, recall, F1 and a full confusion matrix for each fraud type so I can judge real-world performance. • Deployment assets: an inference script or small REST service that accepts an MP3 file and returns the predicted class wi...
...practical implications: consumer fairness, false accusations, privacy, and regulatory context • Limitations and future research • References in the target journal’s citation style Methodology: Conceptual/analytical review with a proposed detection framework. Optionally include an illustrative case study or a small simulated dataset demonstrating one detection approach (e.g., a fake-review classifier or a returns-anomaly model). Note clearly whether any primary data or simulation is expected; otherwise treat as a conceptual review with a proposed framework. Tone & style: Academic, evidence-based, neutral. Original prose only — no plagiarism, no AI-detectable boilerplate, all claims cited. Include an ORCID placeholder for each author. Deliverables: Ed...
I have an anonymised set of internal company emails that must be routed automatically to the correct department—Human Resources, Finance or IT Support. I am looking for an engineer who can take the project from raw data through to a Docker-ised REST service. The workflow I expect is: • Data preparation: strip any residual PII, apply consistent labelling for the three target departments, and document the pipeline so it can be rerun when fresh mail arrives. • Modelling: start with a TF-IDF + SVM baseline, then fine-tune transformer models (BERT or RoBERTa via Hugging Face). Compare approaches and capture precision, recall and F1 for each class. An error analysis explaining common misclassifications is essential. • Deployment: package the best model behind a FastAPI...
I’m looking for an AI engineer who can take ownership of a Natural Language Processing project focused on classifying incoming emails. The end-goal is a robust model that can automatically tag or route messages based on their content so that our internal workflows become faster and more consistent. Here’s what I need: • Data pipeline – guidance on collecting, cleaning, and labeling a representative email dataset while keeping privacy top-of-mind. • Model experimentation – work with modern text-classification approaches (traditional ML baselines through Transformer architectures like BERT or RoBERTa) and justify the final choice with clear metrics. • Training & evaluation – deliver precision, recall, and F1 results on a held-out test se...
...betting model. The focus is behavioural pattern detection and microstructure analysis of bookmaker movement. The framework should build a “corrective lens” per: • league • bookmaker/provider • market type The lens must calibrate automatically from historical data rather than manual weighting. Deliverables: • Cleaned and structured XLS dataset • Rerunnable Python notebook/framework • Signal vs Drift classifier with confidence scoring • League/provider behavioural calibration engine • Visualisations of movement patterns and provider behaviour • Metrics workbook • Future-ready framework capable of recalibrating on new data imports Strong preference for candidates with experience in: • anomaly detection • quantit...
...LoRa mesh network (2–100 nodes) with AES-128-CTR encryption, TDMA-lite slot staggering, runtime role promotion, and RTC-synchronized sleep cycles Seven on-device ML inference models running fully offline on Raspberry Pi: MobileNetV2 (TFLite INT8) for crop disease detection, FAO-56 Penman-Monteith + Random Forest for irrigation planning, Gradient Boosting for yield forecasting, Random Forest classifier for crop recommendation, Isolation Forest for sensor anomaly detection, stage-aware NPK fertilizer scheduling, and a composite Farm-Plan ensemble A three-tier LLM cascade (Anthropic Claude Haiku → Claude Code CLI → Ollama Qwen2.5:0.5B) with silent rule-based fallback for offline resilience A trilingual voice assistant (English / Hindi / Telugu) with offline STT and...
...OCR & Advanced Image Preprocessing Dynamic switching between Tesseract (for speed) and EasyOCR (for complex layouts and handwriting) based on confidence thresholds. Image preprocessing pipeline including auto-skew correction, adaptive thresholding, grayscale conversion, and contrast enhancement (CLAHE) using OpenCV. 3. High-Precision Hybrid Classification Dual-layer classification matching a ML classifier model (scikit-learn) with a weighted keyword registry. Smart disambiguation rules to separate invoices from utility bills (electricity, gas, water, internet, phone, credit cards, insurance). 4. Anti-Hallucination Validation Engine Strict data validation against patterns (e.g., GSTIN, PAN numbers, dates, amounts) defined by regex and business logic rules. An anomaly detection...
...challenges Para 2: CXR as primary tool + radiologist workload/error rates Para 3: Deep learning CNN success + black-box problem in healthcare Para 4: XAI + summary of our proposed multimodal work with 3 contributions 2. SECTION III. METHODOLOGY [Edit ONLY these subheadings] 4.1 System Architecture: Pipeline text + describe Fig 1: Data → Preprocessing → ResNet-50 + ClinicalBERT → Fusion → Classifier → Grad-CAM 4.2 Dataset Description: Create Table 1 like reference - Source: MIMIC-CXR, Pairs: 1,495, Classes: 14, Split: patient-level 70/10/20. Add sentence: "Due to computational constraints, we used a curated subset following standard practice [ref]." 4.3 Data Preprocessing: Bullets - Resizing 224×224, Normalization ImageNe...
I need a compact Power Automate flow that watches my inbox and springs into action the moment a new email arrives. The only emails it should touch are those whose body text matches a classification made by an AI/KI model—think AI Builder text classification or a custom endpoint you connect to with an HTTP action. Workflow is created, i need help by finding out the error or the issue why the prompt output is wrong. Once that positive match comes back, the flow must continue to whatever next step you wire in (for now a simple confirmation email to me is enough; I will extend it later). If the AI says the message is irrelevant, the flow ends quietly. What I expect from you • An exported .zip of the completed cloud flow • A brief read-me showing the connectors used (...
...existing control software. Here is how I picture the workflow: • Data handling: build an ingestion pipeline that pulls real-time feeds (Kafka or MQTT are fine) alongside batch uploads of past performance files, then stores everything in a format that supports fast feature extraction. • Model development: use Python with TensorFlow, PyTorch, or an equivalent deep-learning framework to train a classifier/anomaly detector that flags incipient and critical grid faults. Please include explainability techniques so our operators can trust the alerts. • Deployment: wrap the model in a lightweight REST or gRPC service, complete with health checks and graceful fail-over logic suitable for on-prem or cloud (Docker/Kubernetes). • Testing & metrics: supply unit...
...Natural Language Processing, and Computer Vision. Sessions will be held in real time, so you should be comfortable presenting via Zoom or a similar platform, demonstrating code examples in Jupyter notebooks, answering questions on the spot, and setting short follow-up assignments that reinforce key concepts. I’d like each topic to include practical case studies—think training a simple image classifier, building a text-classification pipeline, or fine-tuning an existing model—so that students finish with portfolio-ready projects. To move forward, please share: • A brief outline or syllabus showing how you would structure • Two sample exercises or mini-projects that highlight your teaching style. • A link to a recent recording (or slide d...
... The project centres on image feature extraction and subsequent classification, so solid experience with OpenCV, scikit-learn or a deep-learning stack such as TensorFlow or PyTorch is essential. You will begin by deciding on (and justifying) an appropriate feature strategy—traditional descriptors like SIFT/ORB, transfer-learning from a CNN, or another proven method—then train and validate a classifier that reaches reliable accuracy on a held-out test set. Clean, well-commented code and clear, reproducible training steps are critical because I need to retrain the model as new data arrives. Deliverables • Python source (scripts or Google Colab file ) covering preprocessing, feature extraction, model training and evaluation • Saved, ready-to-use model weigh...
Current stack: Help Scout OpenAI Google Sheets Later: Shopify Draft Orders What I already have: Make scenario created Help Scout trigger connected OpenAI classifier connected Router with 4 branches: FAQ Quote requests Missing information Supplier requests Google Sheets with: FAQ answers product pricing AI rules What I need help with: Fix Help Scout webhook triggering reliably Configure OpenAI structured output correctly Configure Router filters correctly Connect Google Sheets lookups correctly Create Help Scout draft replies automatically Make the full flow stable and production-ready Test the entire workflow end-to-end Desired workflow: Help Scout email → AI classifies message → Make router selects correct flow → Google Sheets lookup → AI/dynamic draft re...
...Review text / comment Order ID (where available — Swiggy / Zomato) Items ordered (where available) Delivery rating (separate from food rating, where shown) Direct review URL (where available) Step 2 — Classify and Tag Sentiment: Positive / Neutral / Negative — derived from rating and / or text analysis Category tag: Food quality / Delivery / Packaging / Pricing / Service / Other — keyword-based classifier, configurable list Flag urgent reviews: rating ≤ 2 OR text contains keywords like sick, refund, hair, stale, cold, missing, complaint Step 3 — Append to Google Sheet Single master Google Sheet — append only, never overwrite Run Date column at the start of each row Separate worksheet tab per platform OR a single tab with Platform column ...
...Sana. I’m launching an Indian-based brand that celebrates youth, culture, freedom of expression and raw creativity. I need is a logo that captures that spirit in a Bold and vibrant style. I’m picturing a lively mix of bright, energetic hues paired with a memorable symbol and text working together rather than separately, so the name and the icon feel inseparable. I have been quite fascinated by Naive design and maximalism. Need assistance in understanding what would be best for the brand. I will be using the logo across merchandise, social channels and large-format prints, so scalability in Adobe Illustrator or an equivalent vector tool is a must. Please keep colours in RGB and CMYK palettes ready for both digital and print use. Deliverables • Primary logo (...
...text-classification pipeline from scratch to production-ready code. Here’s what I need from you: • Analyse the problem statement I provide and suggest the most suitable model architecture (traditional ML or a transformer-based approach if warranted). • Prepare and clean the text data, engineer any helpful features, and handle class imbalance where necessary. • Train, validate, and fine-tune the classifier, then document the metrics so I can clearly see accuracy, precision, recall, and F1 on a held-out test set. • Package the final model with inference code (Python preferred; PyTorch, TensorFlow, scikit-learn or similar are all acceptable) plus a short README so I can reproduce the results on my machine. Acceptance criteria 1. Reproducible...
I’m building a full, Python-based SMS spam detection module and need a freelancer who can deliver fast. The core of the solution must rely on supervised learning—think Naive Bayes, Logistic Regression, SVM, or any other well-reasoned algorithm you can justify—so that the model can be trained, evaluated, and fine-tuned on a labelled dataset. Here’s what I expect: • Clean, well-commented Python code (preferably in a single Jupyter notebook or a small, logically structured repo) • A documented preprocessing pipeline for text messages (tokenisation, stop-word removal, vectorisation with TF-IDF or similar) • Training, validation, and test results showing key metrics such as accuracy, precision, recall, and F1-score • Short read-...
I will hand you a two-column CSV of real email bodies and their spam/ham labels. From that single file I need a full production-ready pipeline: • Text preprocessing that cleans each message, tokenises it and converts it to TF-IDF vectors. • Training loops for Logistic Regression, Multinomial Naïve Bayes and linear-kernel SVM, with code that automatically picks the best performer. • A FastAPI service exposing /predict so any caller can POST raw email text and receive the predicted class plus a probability score. • A clear evaluation notebook or script that prints accuracy, F1, ROC-AUC and a confusion matrix so I can verify performance. • Repository structured for easy hand-off, including and a concise README explaining setup, training and API u...
...for image preprocessing • API pipeline: o Upload → Image Processing → Feature Extraction → Prediction → Report ________________________________________ 4.2 Face Reading Astrology User Flow: 1. User uploads face image 2. AI analyses: o Face shape o Forehead o Eyes & nose structure 3. Generates personality & destiny insights Tech Stack: • Face Detection: MediaPipe / OpenCV • AI Model: CNN-based classifier • Optional: Emotion detection layer ________________________________________ 4.3 AI Chat Assistant (Optional) • Instant astrology Q&A • Pre-trained astrology dataset • GPT-based or custom LLM integration ________________________________________ 5. Mobile App Features Platforms: • Android (Kotlin / Flutter) &bu...
...trading bot you have built (with execution proof) IBKR-based system you have worked on Screenshots or logs showing real trade execution PnL results or performance data Generic applications, theoretical knowledge, or ChatGPT-generated proposals without proof will not be considered. Current Situation The bot has been built with significant effort and includes: Discord ingestion Trade parser and classifier IBKR execution layer Position management system AI decision layer (Persistent Trader concept) However: Execution is inconsistent Orders fail, hang, or misfire AI layer is not yet reliable System is not safe for real capital In short: everything exists, but it does not work reliably in production. What You Need to Deliver 1. Execution Reliability (Top Priority) 100% reliabl...
...trading bot you have built (with execution proof) IBKR-based system you have worked on Screenshots or logs showing real trade execution PnL results or performance data Generic applications, theoretical knowledge, or ChatGPT-generated proposals without proof will not be considered. Current Situation The bot has been built with significant effort and includes: Discord ingestion Trade parser and classifier IBKR execution layer Position management system AI decision layer (Persistent Trader concept) However: Execution is inconsistent Orders fail, hang, or misfire AI layer is not yet reliable System is not safe for real capital In short: everything exists, but it does not work reliably in production. What You Need to Deliver 1. Execution Reliability (Top Priority) 100% reliabl...
I need a robust, high-performance automation tool to verify and categorize approximately 300,000 user accounts from two databases. The tool will check login validity on a specific portal and classify accounts by type. Key Technical Challenges: Large Scale Processing: The tool must handle 300,000+ entries efficiently. I need a developer who understands multi-threading and asynchronous processing to ensure the task doesn't take weeks. Smart De-duplication: There are many duplicates across the databases. The tool must include a pre-processing step to clean and de-duplicate the list before starting the verification process to save time and resources. Account Classification: For every successful login, the tool must scrape the account dashboard to identify if it is a Corporate (Enterpri...
...include some missing values * include some borderline/noisy cases * include some contradictory cases * have labels generated logically from feature combinations, not randomly Required outputs for the dataset part: * `dataset/` * `dataset/` * `app/ml/` The ML system must be a **decision support system**, not a blind automatic classifier. The intended flow is: * user enters a case * system stores it in MySQL * ML model gives a preliminary prediction * system shows confidence/probabilities * expert reviews it * expert can confirm or change the final classification * system stores both model output and final expert decision * system stores audit history Classification categories: * Suspected homicide * Suspected suicide * Suspected
...include some missing values * include some borderline/noisy cases * include some contradictory cases * have labels generated logically from feature combinations, not randomly Required outputs for the dataset part: * `dataset/` * `dataset/` * `app/ml/` The ML system must be a **decision support system**, not a blind automatic classifier. The intended flow is: * user enters a case * system stores it in MySQL * ML model gives a preliminary prediction * system shows confidence/probabilities * expert reviews it * expert can confirm or change the final classification * system stores both model output and final expert decision * system stores audit history Classification categories: * Suspected homicide * Suspected suicide * Suspected
...information into a reliable attrition-prediction pipeline. Work starts with careful cleaning and preprocessing: handle missing values, encode categorical variables, standardise or normalise where needed and document every step so the workflow is fully reproducible. A brief exploratory analysis should follow to highlight key attrition drivers and verify data quality before modelling. For the classifier, I’d like you to focus on K-Nearest Neighbours. If you find that another algorithm beats KNN convincingly, feel free to present the comparison—but please include KNN in the final report. Train, tune and validate the model, then evaluate it with accuracy, precision, recall, F1 and ROC-AUC. I expect a concise explanation of hyper-parameter choices and cross-validation re...
...Exploratory Data Analysis (EDA) Performed deep EDA to uncover: Customer behavior trends Churn patterns across geography, age, and balance Correlation between features and churn Created visualizations: Heatmaps, distributions, count plots Identified key drivers of churn: Age, inactivity, low engagement, and account balance Machine Learning Models Implemented Logistic Regression Random Forest Classifier K-Nearest Neighbors (KNN) Support Vector Machine (SVM) XGBoost Gradient Boosting Handling Imbalanced Data Applied SMOTE (Synthetic Minority Oversampling Technique) to: Balance churn vs non-churn classes Improve recall and F1 score for minority class Used class weighting for better model fairness Model Performance Summary Evaluated using: Accuracy Recall F1 Score ROC-AUC Score Key...
I have a large, continually growing collection of emails that needs to be processed automatically. The goal is twofold: 1. Classify each email into predefined business categories with high accuracy. 2. Extract relevant entities (names, dates, IDs, product references, etc.) from the same messages. You will own the entire machine-learning workflow. That means cleaning and exploring the raw email text, crafting useful features, training and tuning your models, and packaging the final solution behind an API that I can call from our existing back-end. Python is a must, and I’m comfortable with either TensorFlow or PyTorch for the deep-learning components—use whichever lets you move fastest. Traditional techniques with Scikit-learn are welcome wherever they make sense. Because t...
The project centres on building a production-ready medical image -classification pipeline that leverages modern deep-learning techniques. I have a labelled dataset and need end-to-end code that ingests the text, handles cleaning and tokenisation, and trains an accurate classifier. Python is the preferred language; The preprocessing must involve Quantum computing techniques using Pennylane. PyTorch, TensorFlow or another mainstream framework is fine as long as the solution is reproducible and easy to extend. Key deliverables: • Well-commented source code (data loading, model, training loop, evaluation) • Clear instructions to run training on a fresh machine (README or notebook) • Metrics report showing accuracy, precision, recall and F1 on a held-out set •...
## **Combined Assignment Question** Using the datasets provided (** and **), apply appropriate data mining techniques to perform classification and clustering analysis. --- ### **Part 1: Naïve Bayes Classification ()** Using the cola preference dataset: 1. Apply the **Naïve Bayes method** to classify the 100 customers into: * **Regular** * **Light** cola preference 2. Based on your model, classify the following new customer: * Male, Married, Income = $42,000, Age = 47 * **State whether the customer prefers Regular or Light**, and justify your answer. 3. Evaluate the overall performance of your model: * How accurate is the classification? * Provide an interpretation of the results and any limitations of the model. --- ### **Part 2:
...discovery time by ~40% for stakeholders by replacing static reports with drill-down filters by location, vehicle type, and accident cause. • Applied DAX-based time intelligence to compare YoY fatality trends, revealing a 15% increase in night-time accidents that guided policy recommendations. AI-Enabled Multi-Disease Detection System | Machine Learning | Python • Built a multi-label ML classifier detecting Diabetes and Heart Disease with 87%+ accuracy using Logistic Regression and Random Forest on a 1,000+ patient dataset. • Reduced false negatives by 18% through feature engineering and threshold tuning — critical for early medical diagnosis applications. • Automated the full pipeline: data preprocessing, EDA, model training, and evaluation ...
I need a Natural Language Processing solution that accurately classifies social-media posts into predefined categories. The raw text will be provided in CSV format; it comes directly from public platforms and carries the usual noise—emojis, hashtags, abbreviations, and mixed languages—so an effective preprocessing pipeline is as important as the model itself. Here is how I picture the workflow. • Data handling: robust cleaning, tokenisation, and normalisation that respects emojis and common social-media shorthand. • Model building: a modern text-classification architecture (transformers via HuggingFace, or a lightweight scikit-learn baseline if you can justify comparable performance). • Training & evaluation: use train/validation/test splits and report ...
This task involves a certain level of complexity, as it requires accurately distinguishing between similar sound patterns while ensuring the model is not affected by amplitude variations. In addition, this approach should not rely on amplitude-based features. Instead, it may require advanced techniques such as blind source separation/localization and modern signal processing methods to improve robustness and accuracy. Although feature extraction methods such as MFCC and PCEN can be used, the results may still be influenced by amplitude levels, which can affect the inference accuracy. Therefore, I will assign this project to a suitable person with the required expertise. Thank you.
This quote covers the development of a synthetic ventilator waveform dataset for PVA (Patient-Ventilator Asynchrony) classification research. Deliverables: 1. Synthetic Waveform Generator (Python) - Realistic pressure, flow, and volume waveforms for normal breathing - All target asynchrony...(PNG) - Ready for clinical review and validation 3. Full Documentation - Methodology document explaining each asynchrony model - Mathematical parameters and physiological rationale - Literature references for each waveform type - Step-by-step guide suitable for PhD committee presentation Timeline: 7 days from start Revisions: Up to 2 rounds of adjustments based on feedback Note: Neural network classifier and mobile app prototype are separate phases to be quoted after dataset validation by c...
I’m developing a graduate-level research project that merges smart textiles with security screening: an item of electronic clothing able to detect concealed drugs by combining millimeter-wave sensing with an onboard AI classifier. Where I am right now • Concept development for the millimeter-wave chip placement and antenna layout is underway, but I need an experienced hand to transform these early sketches into a fully realised design and working prototype. What I need from you • End-to-end design and prototyping of the garment, selecting suitable fabrics, conductive threads, flexible PCBs and power management solutions that can live comfortably inside everyday clothing. • Integration of both technologies—millimeter-wave chips fo...
I'm looking to develop an image classifier model to detect and classify patient ventilator asynchrony events, specifically trigger asynchrony, cycle asynchrony, and flow asynchrony. Currently, I don't have any data for training the model. I need to generate synthetic waveforms using a combination of mathematical models and rule-based algorithms. Key Requirements: - Generate synthetic waveforms for training - Use a combination of mathematical models and rule-based algorithms - Classify three types of asynchrony events: trigger, cycle, and flow Ideal Skills and Experience: - Proficiency in Python - Experience with synthetic data generation - Knowledge of image classification and machine learning - Familiarity with ventilator waveforms and asynchrony events
I’m developing a Flutter application that must run completely on the user’s device. Using TensorFlow Lite together with MediaPipe, the app should: • accept images taken directly from the camera or selected from the gallery • perform all processing offline, without any server calls • classify the image, return a set of labels, and generate a short auto-caption in real time I will supply UI mock-ups; what I need from you is the full integration of a suitable TFLite model (or a pair of models, if one is better for captioning) and the MediaPipe image pipeline, plus clean Dart code that exposes a simple method such as classifyImage(File img). Final output should include the Flutter project, the model files, brief setup notes, and a README that explains how to re...
I am looking for help to build a cybersecurity pipeline for automated source code vulnerability analysis. I need an end-to-end architecture that handles detection, localization, explanation, and remediation using a combination of Deep Learning and LLMs, rather than a simple demo. Technical scope: DL-Powered Detection & Localization: A multi-class classifier to categorize multi-language code (starting with C/C++) at the function or file level. It must predict whether code is Safe, belongs to specific top CWE classes, or falls into an "Unknown/Other" category. It must also pinpoint suspicious line numbers and code segments. Code Processing: Use sliding window techniques for long code—no simple truncation. LLM Explanation Generation: An LLM pipeline to output d...
...specifications - Dashboards: KPIs and management panels Technical Requirements Backend (Laravel 12) - Experience: Laravel 12, PHP 8.2+, PostgreSQL - Concepts: Multi-tenant, migrations, observers, events - Integrations: REST APIs, webhooks, OAuth authentication - Standards: PSR, SOLID, clean architecture Frontend (Vue.js/Vben) - Experience: Vue 3, TypeScript, Vite, Pinia - Framework: Vben Admin (Naive UI) - Concepts: Componentization, state management, routing - Standards: Composition API, reactivity, optimization Development - Methodology: Agile, incremental deliveries - Versioning: Git, semantic, code review - Testing: Unit, integration, E2E - Deploy: Docker, CI/CD, cloud environment Benefits and Advantages - Real Project: Production ERP system for public management - Mode...
...prompt to impact. NON-INVASIVE ARCHITECTURE FULL FEATURE LIST 1. LLM Observability Engine What it does: Detects hallucinations, contradictions, ambiguity, or timeouts Back-End: FastAPI, Prompt Log DB, Anomaly Classifier Front-End: Table view, risk icons, filters 2. User Behavior & Friction Analysis What it does: Detects frustration, reprompt loops, abandonment Back-End: Event tracker, heatmap aggregator Front-End: Heatmaps, UX breakdown, friction flags 2.1 Failure Detection & Alert System What it does: Real-time alerting for broken behavior Back-End: Classifier (LLM + rules), notifier API Front-End: Risk graphs, alert logs, admin config 3: AI Ethics & Risk Monitoring (Comprehensive) Multi-layer ethics detection pipeline Toxicity det...
I am ready to dive into natur...exploratory analysis, then walks through feature engineering (tokenisation, embeddings, etc.), model selection, training, evaluation and deployment. • Well-commented Python notebooks and sample datasets so I can reproduce every step on my own machine. • Short explanations of the underlying math concepts, delivered in plain language. • At least one mini-project where we build and benchmark a text-classifier end-to-end. • Live or recorded walkthroughs so I can watch your workflow and ask questions. I learn fastest by doing, so each concept should be paired with code I can immediately run and modify. If this format works for you, let me know how you would structure our sessions and what materials you already have that can a...
