Company name parser and website classification engine

Create a system for human operated web based data extraction and classification. Core to the system is an efficient method for copying Company contact information from the contact page of a website to a database. Each job will also have a few custom questions about the website that the operator can answer. The results get fed into a database. A custom question might be, Does the website have a shopping cart? The company contact data elements of interest are: * Website location where the contact information was found: <[login to view URL]> * Company Name: ACME Manufacturing, Inc. * Company Address: 136 North 42nd Street * Company Suite, box, or Room #: * Company City: Springfield * Company State: OR * Company Zip code: 97478 * Company Country: USA * Company phone: 541-741-2200 * Company Email: [mail [at] [login to view URL]][1] * Company contact: Wayne Van Damme * Company contact phone: 541-554-9414 The goal is to make this process as efficient and accurate as possible for the small team (5-10) of human operators. Please describe your approach here. Manager interface: A Manager can add/delete users and upload lists of URLs and allocate those URLs to users. There will be a management control panel that allows the manager to track progress on a per user basis with some basic user stats such as total number of URLs in queue, total # of completed records submitted, number of URLs reviewed. A manager can load a list of URLs and then assign blocks of URLs to users for contact page parsing (alternatively, users may 'pull' lists once their queue falls below a certain threshold). The manager is responsible for checking accuracy and can flag a user submitted record as accurate or inaccurate. These stats are of course recorded. As stated above, Managers can create questions that must be answered for each record. Each question would have a single field answer that is recorded in the database.