Naive bayes classifierpekerjaan

I have already deployed a full Streamlit application that predicts loan approvals in real time (live demo: , source: ). The pipeline currently includes Logistic Regression, K-Nearest Neighbors, and Naive Bayes models with standard scaling and the usual EDA-driven feature engineering. What I want now is a measurable lift in overall model performance, with the F1-score as the guiding metric. Feel free to explore more advanced algorithms (e.g., Gradient Boosting, XGBoost, LightGBM, calibrated ensembles, or even a tuned version of my existing classifiers) as long as they integrate cleanly with the existing Python | Pandas | NumPy | Scikit-learn stack and can be surfaced through the current Streamlit front-end. Key points you should address •

...working optimiser that reproduces those steps in NumPy/SciPy. • EM for a constrained Gaussian Mixture Model – step-by-step derivation of the E and M updates with the specified covariance constraint, plus a clean implementation that converges on synthetic and real data. • Naive Bayes spam classifier – closed-form derivations for the parameter estimates and a vectorised implementation that processes the provided e-mail corpus. Once the above are working, the same dataset will be used to train and compare Naive Bayes, logistic regression, and K-Nearest Neighbours. I need accuracy, precision/recall, ROC where appropriate, and confusion matrices, followed by: • A 2-component PCA projection with each classifier’s decisi...

...currently at MVP stage and needs to be production-ready before new features are added. --- ## Scope of Work **Phase 1 – Production Readiness** — Review the existing MVP, identify gaps, and ensure the system is stable, reliable, and ready for real user traffic. This includes error handling, edge case coverage, and any necessary refactoring. **Phase 2 – Multi-Product Support** — Refactor the classifier and specification logic to handle multiple products in one conversation, collecting specs for each sequentially without confusing the user. Update the API payload to send a correctly structured multi-product array. **Phase 3 – File Attachment Handling** — Enable file uploads (PDFs, images, drawings) and integrate vision/document parsing so the...

...Sensitivity/Specificity For Segmentation: Dice Score IoU Generate: Confusion matrix ROC curve Document performance clearly. STEP 8: Treatment Prediction Module Once diagnosis model works: Option A: Feature Extraction Remove last classification layer Extract deep features Option B: Combine with Clinical Data Input: CNN features Age Stage Biomarkers Train: Fully connected neural network OR XGBoost classifier OR Survival regression model Output: Treatment response probability STEP 9: Add Explainability Healthcare requires transparency. Implement: Grad-CAM Attention maps Heatmap overlay on PET image Output: Visual tumor highlight Model attention region STEP 10: Backend API Development Using FastAPI: Endpoint 1: Upload PET scan Endpoint 2: Run inference Endpoin...

...right next step automatically. For every Work-related message it should be moved into a dedicated Work folder, flagged for follow-up, and—when I switch the option on—receive a short, templated acknowledgement. Promotions belong in their own folder but still get an auto-reply confirming receipt, while Personal mail is shuffled into its folder and simply flagged so I remember to answer later. The classifier itself can be a lightweight machine-learning or rules-based model; accuracy matters more to me than the particular library, though tools such as scikit-learn, spaCy, or even a fine-tuned transformer are all acceptable. Training data is limited, so please allow for easy retraining or keyword expansion from a JSON or CSV file. The script will connect through IMAP (Gma...

I need an experienced computer-vision developer to build a photo-based image classification pipeline using OpenCV. The system will ingest still photographs taken at live events and automatically tag each shot into predefined categories (for instance crowd, stage, speaker, logo, VIP, etc.). The core requirement is accurate, fast classification of photos only; we are not dealing with video or live camera feeds right now, though I may extend in that direction later. You are free to choose the underlying framework—TensorFlow, PyTorch, scikit-learn—so long as OpenCV is used for image handling and preprocessing. Here is what I expect: • A well-documented training script that reads my labeled dataset, performs augmentation where helpful, and outputs a reproducible model. ...

I’m building a product that relies on fast, accurate text classification and I need a bespoke natural-language-processing algorithm developed from scratch. The goal is to input raw text and have the model return reliable category labels with clear confidence scores. Here’s what I’m expecting from you: • End-to-end code (Python preferred) that trains, validates, and serves the classifier • A well-commented model architecture using mainstream libraries such as PyTorch, TensorFlow, or scikit-learn—whichever best fits the task • Reproducible training pipeline: data pre-processing, tokenisation, hyper-parameter tuning, and evaluation metrics (precision, recall, F1) • A lightweight inference script or API endpoint so the model can slot st...

...touch—yet remain versatile enough for summer dresses and blouse. Here’s what I’m after: the flowers should be PAINTED BY HAND in gouache, watercolour or a similar medium, then delivered as high-resolution (300 dpi) scans. A transparent background or carefully cropped edges will help my print technician drop them straight into our repeat layouts in Photoshop. The flower motifs should be not too naive. They should be modern and sophisticated. Not too commercial and mainstream. The flower motifs will be used in high end / couture fashion. Not budget fashion. The right designs should be seen on a runway. Final files need to be RGB layered PSD or TIFF so we can tweak colors before sending them off to the mill. I am not interested in computer generated ...

I’m building a unsupervised classifier that learns jointly from audio recordings and accompanying physiological signals. My end-goal is a robust prediction model that can generalise to new subjects, so every modelling choice—from feature pipeline through network architecture and hyper-parameter search—has to be evidence-driven and reproducible. Here is what I already have: raw multichannel wave files, synchronised physiological traces (ECG, EDA and respiration) and a draft protocol for train-test splits. What I still need is the deep-learning firepower to turn this into a working model, coded cleanly in Python with TensorFlow or PyTorch, complete with training scripts, inference wrapper and clear documentation. I’ll share the data dictionary, baseline metri...

...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...

I need a researcher who can build a production-ready model that listens to a baby’s cry, watches the paired video, and decides—reliably—whether the cause is hunger, discomfort, or simple attention seeking. Audio and video must be fused inside one architecture; running them in parallel but independently will not satisfy our accuracy goals. You may use the deep-learning stack you trust most (PyTorch, TensorFlow, Keras, OpenCV, torchaudio, etc.) provided the final network can run in real time on an edge device and be exported to ONNX or TFLite. I will share product constraints and a small proprietary data set; you will expand it through public sources or augmentation, perform rigorous cross-validation, and refine the model until we consistently exceed 90 % precision and rec...

I have a curated dataset of abdominal X-ray images that needs a robust deep-learning model capable of classifying key clinical findings. The end goal is a production-ready Python solution that can consistently score above 90 % accuracy on an unseen validation set. You’ll start with any mainstream framework you prefer—TensorFlow, Keras, or PyTorch—and handle the full pipeline: data preparation and augmentation, model architecture selection, training, hyper-parameter tuning, and evaluation. Please keep the code modular and well-commented so I can retrain or fine-tune later as new data comes in. A concise report that explains your decisions, metrics, and suggestions for future improvements will also be appreciated. To help me choose quickly, focus your proposal on your exp...

I have a collection of X-ray studies and I need a robust deep-learning model that can look at each image and instantly tell me which predefined category it belongs to (e.g., chest PA vs. chest lateral, cervical spine, hand, etc.). The job is strictly about classifying the type of X-ray, not diagnosing any pathology. Here is what I already have and what I expect from you: • A curated folder structure with several thousand labelled PNG and DICOM files that you can download from my secure server. • A preference for Python with either PyTorch or TensorFlow/Keras—use whichever framework you feel will achieve the best accuracy and fastest inference on a modern GPU. • Clean, reproducible code (Jupyter notebook or script) plus a short README that explains environment se...

The project centres on building a production-ready text-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; using 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 • Exported model weights and a small inference script or API endpoint for batch pre...

I want to stitch together a fully automated workflow in n8n that is assisted by an AI agent. The core objective is hands-free workflow automation spanning Google Workspace, Salesforce and Slack so I can quit the repetitive busywork and focus on higher-value tasks. Here is the scope I have in mind • Email management – an n8n flow should watch Gmail, classify inbound messages with an AI classifier (OpenAI, LangChain, or your preferred library), file or label them, surface high-priority threads in Slack and, when relevant, create or update Salesforce records. • Data synchronisation – contacts, deals and support tickets must stay in sync between Salesforce and Google Sheets / Drive with conflict resolution rules. • Task management – when certain t...

Project Title: AI-Based "Digital Arrest" Scam Detection System (MVP) Project Overview: I am looking for an AI/ML developer to build a functional prototype of a security system designed to dete...and video data), or will these run as separate independent modules?" Option A: The Screen-Reflection Test Implement a feature where the screen flashes a random color sequence. Build a CV model that attempts to detect this color change in the reflection of the caller's eyes/glasses. Goal: Prove the caller is a live feed and not a deepfake/loop. Option B: Environmental Consistency Check Build a classifier that labels the "Visual Scene" (e.g., Office, Outdoors, Car) and the "Audio Scene" (e.g., Echoey, Windy, Traffic). Trigger an alert if they do not ma...

I need a software solution to streamline property deal information from WhatsApp. Requirements: - Classify incoming messages and images as relevant or junk. - Extract and organize the following property details into a spreadsheet: - From text: Price, Location, Property Type, Sender Details, Size, Plot Number, Block - From images: Text details embedded in the image Ideal Skills & Experience: - Experience with WhatsApp API - Proficiency in image processing and text extraction (OCR) - Strong background in data organization, preferably in spreadsheet formats - Familiarity with classification algorithms and junk mail filtering

...characteristics of the popular website of Mashable (). Hence, this dataset does not share the original content but some statistics associated with it. The original content be publicly accessed and retrieved using the provided urls. All sites and related data were downloaded on January 8, 2015. The estimated relative performance values were estimated by the authors using a Random Forest classifier and a rolling windows as assessment method - see Fernandes et al. (2015) for more details on how the relative performance values were set. The main variable of the study is the number of shares which measures the popularity of the site/post. We are interested to identify the ingredients of a successful post and what it takes to for a post to become a viral. Each student will han...

...summaries and tagged as Positive, Negative, or Neutral. The result I need is a clean JSON output per record, so each review comes back with its summary and sentiment label in a machine-readable format. Because the language is highly nuanced, I’d like you to blend both rule-based and machine-learning techniques: think lexicon cues for idiomatic Telugu alongside a fine-tuned transformer or any other classifier that lifts accuracy. Feel free to draw on pretrained Telugu-BERT, FastText, spaCy, custom dictionaries—whatever combination you believe delivers the most reliable hybrid model. Deliverables • Python or notebook script that ingests raw Telugu text and produces the JSON format • Trained model files (and any custom lexicons) with version control &bu...

I need a machine learning model for text classification tasks. The classifier will be trained to categorize 'controls' data. Requirements: - Develop and train a machine learning model - Perform data preprocessing and feature extraction - Provide clear documentation and usage guidelines Ideal Skills: - Expertise in machine learning algorithms - Proficiency in Python and relevant libraries (e.g., scikit-learn, TensorFlow) - Experience with text data and classification tasks - Strong analytical and problem-solving skills Please share relevant work experience and project examples. Looking forward to your proposals!

4 Milestones - Diagram design - Word-craft (create sketches of words) - Fractal phrasing (sketching and manipulating fractal designs - Final drafts onto master template Instructions to be provided on request, however, see the milestones PDF for a bit more information. Strictly for concept art with pen/pencil/graphics tablet at hand. This requires good sense of science, rationality, arithmetic and English in order to understand the drawing tasks. It should not take more than a few days but I can wait a week. Kindly post regular updates if awarded. If I don't know you and this does not get awarded to someone I already know, send me links to your portfolios. Let me know what you studied, and tell me about recent artwork you have done.

...dependencies light. The key deliverables are: 1. Fully functional one-time payment flow using Stripe. 2. AI-driven categorisation of each successful payment, stored in my data store. 3. Clear, step-by-step setup instructions so I can reproduce the configuration in staging and production. If you have previous examples of pairing Stripe with ML tools like TensorFlow, PyTorch, or even a SaaS NLP classifier, that would be great to see, but I mainly care that the final handoff is clean, tested, and documented....

...algorithm. The strategy must simultaneously cover ten Vanguard ETFs (VIS, VAW, VTWO, VIOO, VTWG, VBK, VIOG, VTWV, VIOV, VFMO) and respect a strict technical rule-set: • Entries fire the moment price touches the 50-day moving average while the RSI confirms healthy momentum. • Exits trigger on a decisive break of the 200-day moving average. • A momentum accelerator and my own “Quantum Edge Meta-Classifier” sit on top to refine every signal. Precision of technical signals, flexibility in position sizing, and a robust audit trail are equally critical; none can be sacrificed. Market regimes (normal, uncertain, stress) must be detected and handled automatically, scaling exposure up or down without manual input. When rates favour value over growth (or vi...

I'm seeking a skilled audio artist to voice and ...series follows a 20-year-old hero and heroine as they navigate murder, mystery, and mayhem in a quaint seaside village where plots include the complexities of criminal and civil law issues. Think: Monk meets Baywatch with a very soft undertone of Christian morality. Key Requirements: - Mixed tone: Dark and suspenseful with light and humorous elements - - Character development: Transformation from naive to seasoned professionals who overcome personal difficulties. Ideal Skills and Experience: - Proven experience in voicing engaging thriller audiobooks - Strong understanding of Maine accents. - - Familiarity with the late 1970s setting and culture, particularly in Maine maybe helpful Please include samples of similar work in...

I need a robust model that can look at a single facial image and tell me, with clear confidence scores, whether it is genuine or a deepfake. The scope is strictly image detection Here is what I expect: • A deep-learning–based classifier trained specifically on faces, capable of flagging “real” versus “fake” with high precision and recall. • A lightweight inference script or REST API endpoint so I can drop an image in and immediately get the authenticity result. • A concise README explaining data preprocessing, model architecture (PyTorch or TensorFlow preferred), and how to reproduce your results. • Evaluation on a well-known benchmark (e.g., FaceForensics++, Celeb-DF) or a comparable dataset we agree on, along with the usu...

...(each of the 16 FSRs may have its own curve/coefficients). The system must support both: A measured interpretation (direct calibrated output). A scaled/normalized interpretation to correct for known recording inconsistencies, explicitly enforcing the constraint 3.00 V @ 450 N when normalization is applied. Calibration alignment must be based on meaningful ramp detection/behavior rather than naive timestamp matching when runs were recorded at different times (consistent with your earlier requirement that “it should fit based on where it sees significant change in the ramp up”). 9) Data output and interface requirements Pico-to-PC streaming: The Pico must stream the complete 16-channel frame in a structured, machine-readable format suitable for real-time parsing (...

...The scope is entirely focused on text data so I’m looking for someone comfortable with modern NLP workflows in Python—think spaCy, NLTK, scikit-learn, or a lightweight TensorFlow/PyTorch setup if you prefer deep-learning. The workflow I have in mind is straightforward: you will start by cleaning and tokenising the texts, engineer any features you deem useful, build and validate the sentiment classifier, then package the finished model with clear usage instructions so I can feed it new text and retrieve the polarity score in one call. Accuracy matters more to me than fancy dashboards, but I do expect a concise README and a notebook or script that reproduces your results end-to-end. Deliverables • Well-commented training script or notebook • Trained se...

...paper writing, just faithful replication and light adaptation. Scope of work • Build a clean, reusable data-preprocessing pipeline for PAN 2015, Pandora and MyPersonality. • Develop the knowledge graph that the original framework relies on, using the sources and schema described in the paper (I will supply all references). • Implement and train the Character-Level Graph Network (CGN)-based classifier within the KE-HHG hierarchy, preferably in PyTorch Geometric or DGL. • Report standard personality-profiling metrics (accuracy, macro-F1, per-trait scores) so results can be compared directly with the published benchmarks. Acceptance criteria 1. Scripts run end-to-end on the three datasets with a single config switch. 2. Model performance reported in...

Sila Dafter atau Log masuk untuk melihat butiran.

...data processing and backend logic, then weave in a supervised-learning classification model, and finally wrap everything in a clean, responsive frontend. Here’s what I have in mind: the core of the system is a well-structured backend that ingests raw data, cleans and stores it efficiently, and exposes clear endpoints. Once that foundation is solid, I want a supervised model—likely a standard classifier such as logistic regression, random forest, or something similarly transparent—trained, evaluated, and seamlessly plugged into the API layer. After that, we’ll add a lightweight UI built with plain HTML/CSS/JavaScript or a modern framework if it speeds things up, keeping the design minimal and easy to navigate. Deliverables • Documented backend code (...

...understands my needs and what you can offer with your pencils, papers, and sharp reading skills. Let me know how busy you are in general. Let me know about your work and what's important for you. What are your hobbies, dreams, and goals. How many years have you been sketching / diagramming? Got loads of screenshots right? (Why not share some links) The subject that I need sketches about is Bayes Theorem. A probability equation. What do you think about it? Here's a video: What kind of sketches would help a layman or 10 year old child understand this? What do you think about testing some example calculations or models, perhaps with Excel? The sketch work, as can be seen in the PDF, can be abstract, it can feature flow charts, pie charts, and

...sensor required) --- Inspection Logic (Phased Approach) I want step-by-step implementation, not complex AI initially. Phase 1 – Rule Based Image comparison with golden reference image Pixel / contour / edge difference Adjustable tolerance Phase 2 – Defect Detection Scratch / dent detection Highlight defect area on image Phase 3 – AI Future Train simple classifier (OK / NOT OK) Dataset will be provided later --- Output & UI Clear result: OK (Green) NOT OK (Red) Display: Live image Captured image Defect highlighted Save: Images of NOT OK parts CSV log (Date, Time, Result) --- Software Behavior Single Python file or small project Software should keep ...

Build a high-performance binary classifier using multimodal data: • images •tabular features The model must incorporate Explainable AI (XAI) In training and using advanced fusion technique.

I need to implement the circuit shown in this paper. Preferably in ltspice or simetrix The circuit is a simple analog based neuron circuit

...is present; if multiple people appear, the SDK must fail fast. • Confirm the person is looking straight into the camera. • Classify and flag: closed eyes, open mouth, face mask, number of detected faces, and overall “live/not-live” status. • Return structured JSON with confidence scores for every rule above so the host app can decide pass/fail thresholds. Performance expectations The classifier should run in real time (≥25 fps) on mid-range devices. A model you have previously trained is preferred, but I’m open to you custom-training or fine-tuning if it increases accuracy, especially for mask and silent-spoof scenarios. Deliverables 1. iOS framework (Swift/Obj-C compatible) and Android AAR, each exposing the same public API. 2. S...

The project centers on building a production-ready TensorFlow 2.x model that classifies tabular data delivered to us through an internal API. I have the API specifications and sample payloads ready; you will turn those streams into a clean training pipeline, engineer the right features, and iterate until the classifier meets our performance targets in real-world tests. Scope of work • Data pipeline – pull the API data, handle preprocessing, and produce TensorFlow-friendly datasets for train/val/test splits. • Model development – design, train, and tune a deep learning architecture suitable for tabular inputs (e.g., wide & deep, Transformer, or other proven structures). • Optimization – experiment with hyperparameters, regularization, and c...

The project centers on building a production-ready TensorFlow 2.x model that classifies tabular data delivered to us through an internal API. I have the API specifications and sample payloads ready; you will turn those streams into a clean training pipeline, engineer the right features, and iterate until the classifier meets our performance targets in real-world tests. Scope of work • Data pipeline – pull the API data, handle preprocessing, and produce TensorFlow-friendly datasets for train/val/test splits. • Model development – design, train, and tune a deep learning architecture suitable for tabular inputs (e.g., wide & deep, Transformer, or other proven structures). • Optimization – experiment with hyperparameters, regularization, and c...

The project centers on building a production-ready TensorFlow 2.x model that classifies tabular data delivered to us through an internal API. I have the API specifications and sample payloads ready; you will turn those streams into a clean training pipeline, engineer the right features, and iterate until the classifier meets our performance targets in real-world tests. Scope of work • Data pipeline – pull the API data, handle preprocessing, and produce TensorFlow-friendly datasets for train/val/test splits. • Model development – design, train, and tune a deep learning architecture suitable for tabular inputs (e.g., wide & deep, Transformer, or other proven structures). • Optimization – experiment with hyperparameters, regularization, and c...

Rationality - Woman's Face & Neckless, Cat, and Thomas Bayes. Project for Elena B. For this task, broken down into 3 milestones I need a few details added to the drawing. Half of the job this time will be to use slightly more realistic techniques as these involve 'real' characters in the scene. 1) - Paint the white Persian cat as described in the PDF. Will require 4 draft sketches. 2) - Paint the woman's face, hair, and necklace. Will require 3 draft face sketches. 3) - Use the standard line art style to draw Thomas Bayes on the panel as described in the PDF.

The project centers on building a production-ready TensorFlow 2.x model that classifies tabular data delivered to us through an internal API. I have the API specifications and sample payloads ready; you will turn those streams into a clean training pipeline, engineer the right features, and iterate until the classifier meets our performance targets in real-world tests. Scope of work • Data pipeline – pull the API data, handle preprocessing, and produce TensorFlow-friendly datasets for train/val/test splits. • Model development – design, train, and tune a deep learning architecture suitable for tabular inputs (e.g., wide & deep, Transformer, or other proven structures). • Optimization – experiment with hyperparameters, regularization, and c...

...end-to-end, live face-recognition model that runs smoothly on Windows and authenticates users from a webcam feed in real time. The pipeline must follow the architecture I already have in mind: • Feature extraction: implement Global Search ShuffleNet coupled with a Generative Adversarial Network (GSS-GAN) from scratch or by extending public research code. • Face cognition / matching: build the classifier with a Convolutional Neural Network optimised for low latency. The model should open a webcam stream, detect a face, apply GSS-GAN for robust feature vectors, and pass them through the CNN to decide whether the face belongs to an enrolled user. An accuracy benchmark on a small hold-out set is fine for now, but the live demo has to stay above 25 fps on a mid-ran...

...long as the end result is accurate and reproducible. I will supply a representative sample of matches for training and evaluation, and can label additional clips if the model needs more data. The system should ingest standard MP4 files, and produce: Build a detection and classification pipeline using: • Roboflow + YOLO, or • Ultralytics YOLOv8/YOLO11 + MediaPipe, or • MoveNet/SensiAI + classifier • Detect: player, racket, ball, pose, shot type. • Compute timing and technical metrics. • Generate structured JSON: "type_of_shot": "bandeja", "strengths": [], "improvements": [], "score": 82, "overlay_url": "" • Generate human-like feedback using GPT-...

I’ve got a collection of time-stamped web server logs and I want to squeeze two clear outcomes from them: 1. A reliable time-series model that forecasts our revenue day-to-day (and ideally beyond) so we can plan inventory and campaigns with confidence. 2. A companion classifier that flags and categorises IT events hidden in the same log stream—anything from routine spikes to anomalies that hint at trouble—so operations can react before customers notice. The data is already centralised; you’ll receive the raw log files plus a cleaned-up sample to speed exploration. I’m open to the modelling stack you prefer—Python with Prophet, ARIMA, LSTM, or even Facebook’s NeuralProphet are fine—as long as the forecasts are explainable and the ev...

...download AUTSL, isolate the RGB stream for every clip, then extract frame-level hand-body keypoints with MediaPipe (OpenPose is fine if you prefer). • Dual-branch network – an RGB pathway built around a 3D-ResNet (or a comparable spatiotemporal CNN) and a Skeleton pathway driven by either LSTM layers or a Temporal Convolutional stack. • Mid-level fusion – combine the two streams before the classifier so they jointly vote on the final sign. • Robustness enhancement – implement Modality Dropout during training to simulate missing channels and toughen the model against scenarios where keypoints fail or footage is blurry. • Evaluation – report Accuracy clearly; you can mention other metrics in logs, but Accuracy is the headline figu...

I have a 1,970-word social-science ...keeps every argument, citation, and heading exactly where they belong while making the prose feel unmistakably human. Smooth awkward phrasing, shuffle sentences when it clarifies flow, vary rhythm, and weave in subtler lexical choices—but do not add new data, change the thesis, or let the word count creep above the original. The finished draft must slip past both GPTZero and OpenAI’s AI Text Classifier with less than 10 % AI probability. I’ll run those checks the moment the file lands in my inbox; if either detector scores higher, I’ll send it back for a quick round of tweaks. Deliverables • A clean .docx with all edits accepted • A tracked-changes .docx showing every modification Return both files ...

...com/code/masahirogotoh/chb-mit-eeg-dataset-seizure-detection-demo). Before the deep-learning section I want a solid, publishable channel-reduction stage that genuinely boosts accuracy, so that fewer electrodes are needed without sacrificing performance. So far I have implemented four recent meta-heuristics—GASO, EVO, Hippopotamus Optimization and the Botox Optimization Algorithm—using a Random Forest classifier as the fitness evaluator (fitness = classification accuracy on the seizure task). Results are promising, but I need a specialist to refine and stabilise this optimisation block and add a clear element of novelty. Key tasks • Convert the current continuous search space to an effective binary representation, OR introduce an Opposition-Based Learning sc...

...analyzing sports videos and generating automated performance feedback. The system should detect and track multiple objects including the player, racket, and ball, as well as estimate body pose and identify the type of shot being performed. The model can be built using one of the following approaches: Roboflow + YOLO Ultralytics YOLOv8/YOLO11 with MediaPipe MoveNet/SensiAI combined with a custom classifier Using the detections, the system must calculate timing and technical performance metrics and output structured JSON in the following format: { "type_of_shot": "bandeja", "strengths": [], "improvements": [], "score": 82, "overlay_url": "" } The solution should also use GPT-4o to generate natural...

Build a lightweight AI tool that reads customer support emails and auto-tags them by category and urgency. The goal is faster triage and routing for our support team. I need a simple AI tool that reads customer support emails and automatically categorizes them into groups like billing questions, technical issues, cancellation requests, or general feedback. It should also flag urgency level such as urgent versus normal priority. The tool should take email text as input and output category label, confidence score, and optional urgency flag. Would be helpful if it can connect to Gmail or IMAP to pull emails automatically and log results to Google Sheets or CSV format for tracking. What I need delivered: - Working prototype that can classify emails - Sample data with README documentation - Opt...

...how attitudes expressed in project documents can feed back into overall performance metrics. Scope • Focus on management challenges within the energy or construction sector (I’m flexible here as long as the management angle is clear). • Work exclusively with an openly available dataset—press releases, project reports, stakeholder comments, or similar text sources. • Build a sentiment classifier (Python, scikit-learn / PyTorch / TensorFlow—your call) and connect its output to a causal-loop or stock-and-flow model created in Vensim, AnyLogic, or a Python equivalent. The dynamic model should illustrate how changing sentiment influences key project-performance variables. Deliverables 1. Cleaned and documented public dataset with acquisit...

I have a small, curated dataset of European songs and need a clear, reproducible demonstration model built in Python using TensorFlow/Keras. The goal is simply to show how we could predict or classify which tracks are most likely to resonate with European listeners—nothing production-grade, just a clean proof of concept that I can study and rerun. Here’s what I’m after: • A short, well-commented notebook or script that loads the data, performs any essential preprocessing, trains a straightforward model, and prints basic evaluation metrics. • Clear instructions (README or inline notes) so I can execute everything on my machine with a fresh virtual environment. • A brief write-up—one pager or a few slides—summarising feature choices, model ar...