4
$\begingroup$

I would like to develop a platform in which people will write text and upload images. I am going to use Google API to classify the text and extract from the image all kinds of metadata. In the end, I am going to have a lot of text which describes the content (text and images). Later, I would like to show my users related posts (that is, similar posts, from the content point of view).

What is the most ppropriate way of doing this? I am not an AI expert and the best approach from my prescriptive it to have some tools, like google API or Apache Lucene search engine, which can hide the details of how this is done.

Improve this question
$\endgroup$

2 Answers 2

Reset to default
1
$\begingroup$

I would suggest to convert the documents into TF-IDF(use Gensim) vectors and then compare them using various similarity calculating techniques like cosine similarity.

You should read this amazing article for the same. I once used it while working on my project.

https://medium.com/@adriensieg/text-similarities-da019229c894

Improve this answer
$\endgroup$
0
$\begingroup$

Google has introduced Universal Sentence Encoder, which converts sentences into vector representations while preserving the semantic details. The pre-trained models are available on Tensorflow Hub. The Colab notebook would help you get started as well.

Improve this answer
$\endgroup$
0

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .