INFORMATION EXTRACTION AS A SERVICE
NERILY is an information extraction service that extracts vital text snipplets from unstructured text. Our service allow our users to entirely automate their menial processes of sieving relevant information of interests from an unlimited set of arbitrary documents.
In the next section, we will demonstrate how NERILY can be used.
In this simple demo, we will use the NERILY service to extract elements of data from a text chunk that you can provide.
For this demo, we are using an engine that is trained with a model that is able to identify and extract entities that belong under the following classifications: Location, Time, Person, Organisation, Money, Percent and Date.
To begin, simply enter an english paragraph and then press Enter.
We built this engine for ourselves to identify semantic entities pertinent to the physical world for when we crawl the web. The easiest way for you to the utility of NERILY is to customise the engine for your exact needs. So we are sharing some of our tools for you to very easily do it for yourself!
1. Download NERILY Chrome Extension
2. Begin training your very own training model using the extension.
3. Generate and provide the training model
4. Enter a chunk of text that contain an entity that you had trained.
Click here for more detailed instructions.
For the demo to work well enough, you need a training set that covers a large enough data set.
NERILY is essentially offering the technology of Named Entity Recognition (NER) as a service. We employ machine learning and natural language processing (NLP) in identifying atomic elements of data via differentiating features that the platform identifies from a large-enough training model. Our engine is trained primarily from a statistical learning approach, hence the larger the training modelset, the higher precision we have with the information extraction.
Our goal is to abstract the technical complexity of NER so you can focus on the utilty implementation in your use-cases. We are able to do so by providing the service as an API and webapp, all while providing the necessary toolkits for you to continually improve on your training modelset so you can achieve a higher precision on the engine.
+1000% IN PRODUCTIVITY
NERILY has enabled entire automation in Spawt's semantic web crawler, regardless of the web structure. With NERILY, Spawt is now able to identify physical addresses, phone numbers, ratings, etc, from any crawling results without any human intervention.
+100% IN PRODUCTIVITY
A teacher spends an average of 11 hours a week outside of curriculum time grading assignments. NERILY is able to efficiate this process by at least 100% by quickly identifying the keypoints that an assignment contain.Case study
+150% IN PRODUCTIVITY
The human role of the auditor is traditionally the largest cost to an accountancy firm. Automation of information retrieval off statements are able to raise productivity of any accountacy firms.Case study
Every user will be provided tools that will allow him/her to instantly employ the information extraction technology in his/her use-cases. These tools include
1. Chrome Extension - for building a training modelset
2. Access to NERILY webapp - for instant access to NERILY's information extraction technology
3. Access to NERILY API - implementation with NERILY technology for your automation use-cases
AS-YOU-GO Plan: USD$19 will net you $20 of credited value. Each document parsed costs $0.01 credit.
COMFORT Plan: USD$49 for a monthly unlimited plan.
LIFETIME Plan: USD$1099 for a lifetime unlimited plan.
Terms and conditions apply.
We stand by our goal of abstracting complexity of information extraction, this is why we want to work closely with your organisation in customising tools and deployment specific for your needs, especially in modelset training. Email us today or call +65 92205385 to set up a meeting with Steven.
We understand the frustrations with bad support. This is why we want to deliver the best support as we best as we can possibly sustain.
If you have questions at all, feel free to email us at firstname.lastname@example.org and we will try to reply as soon as we can.
Otherwise, you too can reach Steven at +65 92205385.