We are digging websites and cloud services of your choice and internal systems for the information and the data you need, with the help of our powerful crawlers and connectors, that are fed with your business rules.
What is it?
Our easy-to-use Data Access Automation Platform will enable you to easily access and analyze the data and information that matters to you and your business.
Use Case : Semantic Price Monitoring
Products and Brands
Directly through an API
The application uses a Product and Brand dictionary listing any single product that the application should track. This is provided by the customer. His knowledge base. The applications also uses offers identified either manually or automatically on websites of online merchants, retailers and marketplaces.
- A catalog synchronisation component in charge taking into account catalog updates (insertion of new product references, updating existing ones, removal is rare but supported)
- A set of per website connectors in charge applying manual extraction rules written for the purpose of the solution
- A web crawler
- An offer identification module able to identify saler offerings on any crawled webpage
- A reconciliation method that is used to perform the alignment of data coming from per website connectors and the automatic wine offer identification module plus the web crawler with the reference catalog provided by the customer
An API in charge of serving the content. Several access means are provided to fit the customer requirements. They usually consists of data listings, either focused on a given product reference, on a given store reference, or a combination of both. The application then spills records accordingly.
Use Case : GeoDataAPI
POIs, Communications, Persons
Directly through an API, through a third party native or web application, or through hosted a web application (TBD, aquitaine.geodataapi.com)
The application uses a mix of Open Data, Crawled web data, and data available from third partis.
- An OpenData resource named OpenStreet Maps to establish a referential of POIs. The dataset is regularly updated. We work with snapshots that address parts of whole dataset (wether it’s by continents, countries or regions of the world). This gives a basis for search POIs.
- A set of connectors that are used to produce additional metadata, organised in “layers” that are then added to POIs existing in our referential. Such connectors / layer provider are : Wikipedia, Google places, Twitter / Foursquare / Facebook, Trip Advisor, and so on. Communications and Persons come from social networks only.
- A reconciliation method that is used to perform the alignment of data coming from connectors with the POIs found in the geographic referential
An API in charge of serving the content. Several modules are involved, such as:
- Access by tags / categories, for POIs exploration puproses
- Geosearch: center+radius, bounding boxes, envelopes, geoshape, …
- Name search: full text search over the name and key metadata of each POI
- Search can also be performed by combining any of the three methods above
- Results listing can be sorted by distance if a reference POIs or coordinates pair is provided, by relevancy, or any metadata associated to the POIs (popularity, review rankings, number of checkins, …)
- A given POI can also be served by ID, with the possibility to filter through the different layers available (to return only the Wikipedia description, trip advisor reviews, google places opening hours, and so on.