Lec 4.3 - Named Entity Extraction Lexalytics

Download PDf directly

Or Read Directly

Named Entity Extraction | Lexalytics What is Entity Extraction? Entities are the who (and some of the what) of text analytics. On the most basic level, an entity in text is simply a proper noun such as a person, place, or product: John Coltrane, Coca Cola, and Indiana are all entities. But Lexalytics doesn’t restrict our analysis to common nouns: several of our entity extraction techniques allow for full customization, so you can create your own definition of “entity”. Below you will find a high-level overview of entities and Lexalytics’ entity extraction functions. For a more in-depth explanation of our named entity extraction, read through “Entities: Who Are They Talking About?”. Check out our web demo to see Lexalytics in action, or get in touch to schedule a live demo. Now that you’ve prepared the text, you can do things like extract the entities, and get the associated sentiment, themes, and summary (for that entity). Named Entity Extraction Lexalytics’ named entity extraction feature automatically pulls proper nouns from text and determines their sentiment from the document. Salience Engine and Semantria Excel/API all come with lists of pre-installed entities and pre-trained machine learning models so that you can get started immediately. People, places, dates, companies, products, jobs, and titles are all automatically detected. Our machine learning models are so sophisticated, you’ll discover entities and relationships you didn’t know about. You'll discover new competitors just entering the market, track the activity of spokespeople at your competitors and customers, and catch new products at the moment of launch. These are just a few ways our named entity extraction tools will improve your visibility into your business landscape. Custom Entities Using our easy online configuration tool, Semantria Web, you can also build your own lists of custom entities for tracking or train your own machine learning models for discovery purposes, to find entities specific to you. Cuts of lumber, types of cancer, variants of a stereo model – anything that your business considers an entity — can be identified and tagged as such. When you feed your fledgling model training content, it will quickly understand the broader category you’re interested in and learn to pick up entities you wouldn’t think to include in your lists. We have trained a model to recognize people, places, companies, and products. We also recognize @handles and #hashtags, as well as other things like dates, job titles, currency amounts. When consumers are saying bad things about your competitors’ products, hit them with targeted advertising to sway them to your side before your competitor even knows there was a problem. If a customer has a bad experience with one of your products, identify the source of their troubles and respond appropriately, building good will and brand loyalty. This just gives you an idea of the things that we can tell you about an entity. You can check our documentation for a more extensive list. Named entity extraction gives you insight about what people are saying about your company and — perhaps more importantly — your competitors. Knowing who is speaking and what they are talking about, and the context which they are speaking in, gives you that critical edge over your uninformed competition.