Find Jobs
Hire Freelancers

A Natural Language Classifier (Machine Learning) to detect spam - 03/05/2018 00:28 EDT

$1500-3000 USD

Cerrado
Publicado hace casi 6 años

$1500-3000 USD

Pagado a la entrega
We need a Natural Language Classifier which will work in as many different languages as possible which can be used to detect spam messages in Push notification of up to 144 characters in length. To train the model, I should be able to upload a .csv file which contains two columns: a sample text and the classification (similar to IBM Watson NLC). Although the model will primarily be used to detect spam, it should also be able to detect other classifications of messages. Example of classifications include: notifications, reminders, phishing, advert-spam, scam(eg. nigerian prince) and others. Not to worry, we have the dataset needed to train the model (3000+ records). After the model is trained, I should be able to query the model using a REST API. Each prediction should take less than 80 ms (the faster the better). In order to achieve these speeds, Any technologies that will work with AWS Lambda are acceptable. Accuracy is extremely important. Please note, I should be able to use open source tools for the entire project. Because we will be training the model on ECS, I should be able to put the entire training process able to put into a Dockerfile. We understand if you don't have the AWS skills(we can hire someone else for that), but it must be compatible with AWS Lambda so that we will be able to put it on there. It's our hope we should be able to use AWS ECS to train the model and put the model in a S3 bucket, and then use AWS Lambda for inference/predictions checks. Technology stack used must be compatible with AWS Lambda. If you are bidding, please declare your stack choice. Also indicate if you have the AWS skills necessary to do this (not needed). Finally please indicate if you have any experience with building Natural Language Understanding APIs. Please also indicate which machine learning algorithm do you feel could be a viable candidate for this project (used to weed out those who don't have Natural Language Understanding experience). Hint: Naive Bayes is a poor choice.
ID del proyecto: 16890536

Información sobre el proyecto

9 propuestas
Proyecto remoto
Activo hace 6 años

¿Buscas ganar dinero?

Beneficios de presentar ofertas en Freelancer

Fija tu plazo y presupuesto
Cobra por tu trabajo
Describe tu propuesta
Es gratis registrarse y presentar ofertas en los trabajos
9 freelancers están ofertando un promedio de $2.569 USD por este trabajo
Avatar del usuario
Experienced data scientist who has extensively worked on NLP/NLU projects using ML algorithms (python). Have used AWS lambda before. Regarding the choice of algorithm we'll have to do some experiments with the training dataset previously have done NLP tasks with LSTM, LSTM-CNN, BoW - SVM etc. For creating the model we'll use Keras/ Tensorflow for NN and Scikit-learn for other SVM/linear/ decision tree based algorithms. Regarding experience in building NLU APIs, we have created APIs for ML models (for computer vision projects) though not for NLU specifically.
$2.200 USD en 30 días
5,0 (50 comentarios)
6,7
6,7
Avatar del usuario
I am a professional ML/NLP Engineer in an IT company with 2+ years of Machine learning and Natural Language processing, have experience working with many challenging problems and also with different tools and libraries like Scikit-learn, keras with backend tensorflow, and word2vec, fasttext, textacy etc, let me know if you are interested
$3.000 USD en 30 días
4,8 (48 comentarios)
6,3
6,3
Avatar del usuario
Hi. My name is Uzair.I am in final semester of my masters in Electrical Engineering. I am doing my thesis in biomedical signal processing and Machine learning. I have more than 3 years of experience in MATLAB specially in Machine learning, Data processing and Signal processing. I also have experience in PYTHON, C++ and C. Due to my vast coding experience, I can handle your project with quality work. Regards
$1.666 USD en 30 días
4,9 (49 comentarios)
5,9
5,9
Avatar del usuario
Hi I would use Python as a language and Keras as a library. It is compatible for AWS Lambda as far as I know. I have experience with Ambiverse and Watson, but I didn't take part to develop any of natural language understanding APIs. I just used them to develop a product. As for algorithm, naive Bayes is really poor choice. Therefore I can develop specific algorithm using SVM, Probabilistic latent semantic analysis and Perceptron. As far as tested it is better than naive Bayes, although not as good as Watson. Thanks Archil
$2.500 USD en 30 días
4,0 (1 comentario)
3,1
3,1
Avatar del usuario
We are team of Scientists and Developers having rich experience with Artificial Intelligence Techniques and Machine Learning Techniques like Neural and NLP. We have done extensive research on Facial Recognition and Computer Vision & Image Processing. The Technical team consists of Programmers having experience more than 6 years.
$4.110 USD en 8 días
5,0 (1 comentario)
1,4
1,4
Avatar del usuario
Hi there, I have read your requirement and I am best fit for this. I am experienced with IBM Watson, Google Tensorflow, Apple Corevision Framework & CoreML and Expert in training ML Model using Python programming. I do have full experience with Caffe and Keras. CoreML , tensor flow. Also, I do have good understanding with Natural language processing and machine learning. Also having AWS expereice with 5+ years with AWS services, EC2, S3 bucket, Lambda functions, Step functions, DynamoDB , Amazon Recognition - Face recognize API and AWS cloud platform , Elastic Container Service for Kubernetes , Machine Learning , Alexa skills set, AWS iOT, Cognito. It would be best to discuss here as I need more details. Thanks.
$2.600 USD en 30 días
0,0 (0 comentarios)
0,0
0,0
Avatar del usuario
We can talk about the details over the chat
$2.444 USD en 20 días
0,0 (0 comentarios)
0,0
0,0

Sobre este cliente

Bandera de UNITED STATES
Canton, United States
5,0
6
Forma de pago verificada
Miembro desde ene 7, 2018

Verificación del cliente

¡Gracias! Te hemos enviado un enlace para reclamar tu crédito gratuito.
Algo salió mal al enviar tu correo electrónico. Por favor, intenta de nuevo.
Usuarios registrados Total de empleos publicados
Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)
Copyright © 2024 Freelancer Technology Pty Limited (ACN 142 189 759)
Cargando visualización previa
Permiso concedido para Geolocalización.
Tu sesión de acceso ha expirado y has sido desconectado. Por favor, inica sesión nuevamente.