Need to create an aggregator with AI components
300 USD So I need a website that will collect listings from other websites (about 6-7 of them) and then identify duplicate listings among all of them, aggregating them into a single listing. Basically, an aggregator, can you do this? It should be accessible via API - providing a list of all unique listings and for each listing - details such as how many duplicates it has and where these duplicates are from (showing a link for each source where the listing was taken from), also providing brief data on each source for the duplicate listing (for example, phone numbers, price, description if it differs, etc.)
Can you do this?
No interface is needed, just an API and an admin panel where you can view the collection results and, if necessary, edit them if the model did not work correctly. If the model can be further trained on this, that would be excellent.
Client's review of cooperation with Georgii Popov
Need to create an aggregator with AI componentsI can't evaluate, but the fact is that the deadline was extended, in the end I don't know what I received, the person says his wife is on maternity leave, I have nothing against it, but the work needs to be done.. and try to complete it on time..
-
14 days500 USD
2754 42 1 14 days500 USDHello, Viktor
Can I review the donor websites?
I have experience in developing APIs and web projects on cross-platform .NET
Maybe I can be helpful..
-
7 days700 USD
1591 8 0 7 days700 USDHello, Viktor
I am a developer in the field of ML/DL. Ready to complete your project. Write to me, let's discuss.
-
5 days400 USD
2491 37 0 5 days400 USDHello!
I will gladly take on your project!
Have extensive experience in this.
Write in private messages and check the portfolio
-
Winning proposal5 days300 USD
1 2 3 Winning proposal5 days300 USDHello! I understand the task perfectly: I have already created parsers that merge duplicate listings based on different features (price, phones, texts, photos). I can build an aggregator with an API, admin panel, and logic for merging duplicates with source attribution. I have experience in machine learning and web scraping, and we can set up model training on examples and provide an editing interface. Implementation — fast, precise, focused on results. If you're available — let's discuss the details.
-
3 days400 USD
99 3 days400 USDYes, I can handle this. You need an internal system that extracts lists from multiple sources, identifies which ones are the same, and clearly organizes them using an API. I have done similar work when data from different places needed to be grouped, cleaned, and made accessible with a simple structure.
For the collection part, I will set up scrapers or use available channels depending on each website. Once the data is entered, I can create an intelligent matching system that detects duplicates even if the wording or price slightly differs. I can use a combination of rules and AI models to make this matching as accurate as possible. Each listing will include the number of duplicates, links to each original version, and comparisons of things like phone numbers, price, or description.
The API will return a list of all unique listings and provide full information when needed. I will also create an admin panel where you can view the collected data, correct anything that seems wrong, and monitor how the system works. If you want, I can make the system learn from your corrections and gradually improve over time.
Let me know when you want to dive into the details. I can sketch out a diagram and develop a plan that matches how you want to proceed.
-
2 days200 USD
1117 4 0 2 days200 USDHello,
It seems that this is something I can do well. I have created several systems that extract structured and unstructured data, clean it, and make it usable through an API. Duplicate detection using AI is quite feasible and can be improved over time if we gather examples where the model made mistakes.
Regarding collection, I would write smart scrapers that can handle changes in layout and structure. Then I would use a combination of text similarity, phone and price matching, and possibly embedding if the content is diverse. Each unique listing will include the original links and detailed differences, such as price or description, if available.
The API will return clean structured data, and I will also provide a small admin panel where you can see what has been collected, check how duplicates were processed, and correct any errors. This feedback can be fed back into the model if you wish.
Let me know if you want to review a sample of data or a small prototype.
-
4 days450 USD
631 5 0 4 days450 USDGood afternoon
I will gladly help you with the implementation of the aggregator creation task
--------
Бишкек