My AI model should do:

Описание вопроса 1

Predict a binary outcome

Classification. Used for: Spam filtering, language detection, a search of similar documents, sentiment analysis, recognition of handwritten characters, and fraud detection

Predict a numeric value

Regression. Used for: Forecasting stock prices, predicting sales volume, etc

Group similar things together

Clustering (Classical ML). Popular use case: For customer segmentation, labeling data, detecting anomalous behavior, etc

Infer patterns (associations) in data

Association Rule (Classical ML). Popular use cases: Helping stores cross-sell products, uncovering how items are related or complementary, and understanding which symptoms are likely to co-occur in a patient (comorbidity)

To generalize data and distill the relevant information

Dimensionality Reduction (Classical ML). Popular use-cases: Recommender systems, topic modeling, modeling semantics, document search, face recognition, and anomaly detection

Other

The number of databases systems that need to be linked and processed

All data is already collected in one database

Need to merge 2 databases

It is necessary to combine 3-5 databases

I have more than 5 databases

Database

We have tabular databases (SQL)

We use noSQL

We use both types: SQL and noSQL

Linking with third-party databases

The algorithm will only use the data that we already have

You need to collect data using the API

Need to collect data using scraping (I have no API)

I have a database in tabular format

You are looking for

Modify the existing algorithm

Create a new model (create a completely new algorithm)

Visualization

We want to present data on our platform / application

We want to get visualization on a separate dashboard

    There are mere trifles left.
    Leave your details and wishes, if you have any, and we will contact you within 1 hour with ready-made proposals for our future project