SEMANTIC SEARCH FOR TRAVEL PLACE DOCUMENT

This project aims at developing a Semantic Searching Engine that could search with meaning not only to find keywords but to determine the intent and contextual meaning of the input sentence. We will demonstrate another finding technology known as Semantic search by utilizing sophisticated Artificial Intelligent technologies and some Natural Language Processing models.

semantic search

INTRODUCTION

intro-1

There are so many searching engines currently, but most of them are using the Keyword Search method. Keyword search is looking at terms wherever they appear, even if part of a larger phrase or used in a different context, Keyword Search will find words anywhere in the record. So the Keyword Search has a limitation in that the words often have multiple meanings, so Keyword Search will return the irrelevant results.

intro-3

Unlike Keyword Search, Semantic Search is a search with meaning, it will try to get at the intent and contextual meaning of terms and your search query. And we have built a Semantic Search engine on the database from https://wikitravel.org/en/Main_Page. It will find and show you the top k (default k=10) articles with the best matching meaning to your search query.

APPROACHS

Data

We crawled all existing articles from https://wikitravel.org/en/Main_Page. After cleaning and preprocessing the crawled data to remove the non-existent articles, wrong, and repeat posts… the final number of articles is 31249 articles. And we store them into a .csv file for use in the searching step.

Text similar model

In order to search the input sentence based on its matching meaning with the article database. We have applied a Deep Learning model to represent the sentence with an embedding vector, and then we will compute the sentence similarly based on these embeddings vectors to find the best matching results.
semantic search-approach

USAGE

semantic search usage-1

Step 01

Access to the Semantic Search for Travel Place site: https://experiment.saigontechnology.vn/travel-search/. Or you can access the main Saigon Technology AI Research Lab page here: https://experiment.saigontechnology.vn/, select the Semantic Search for Travel Place section and click Try our demo button.

semantic search usage-2

Step 02

On the Semantic Search for Travel Place page, you could see a UI like this.

semantic search usage-3

Step 03

Type the sentence you want to search in the text input area.

semantic search usage-4

Step 04

Then click the Search button to start searching on the wikitravel database.

semantic search usage-5

Step 05

The top 10 best searching results will be displayed as bellow:

Next Case Studies

The Natural Language Processing Toolkit (NLTK) is a Python-based software application that offers a suite of tools for the purpose of processing natural language data.
Utilizing AI-based Computer Vision techniques, the Product Recognition system autonomously detects and categorizes products present within images or videos.

Let’s Talk

Together with our developers and analysts, we begin by discussing and analysing our client’s needs, sketching the outline