Last week my team ‘wrote’ 12 million words of professional narrative about real estate professionals. Each ‘story’ was unique, free of grammatical errors, optimized for search and highly effective.
How did we do this? By combining artificial intelligence and natural language processing.
In this article, I hope to explain what is artificial intelligence and natural language processing and how it can create great content for real estate professionals.
What is artificial intelligence? Artificial Intelligence, or AI, means making machines capable of performing intelligent tasks like human beings. AI performs automated tasks using intelligence.
The term Artificial Intelligence has two key components:
There are different levels, or stages, of AI.
- Stage 1 – Machine Learning – It is a set of algorithms used by intelligent systems to learn from experience.
- Stage 2 – Machine Intelligence – These are the advanced set of algorithms used by machines to learn from experience.
- Stage 3 – Machine Consciousness – It is self-learning from experience without the need of external data.
What is Natural Language Processing (NLP)? NLP is a subset of AI; it means making machines capable of understanding and interpreting human written and spoken language. The objective of NLP is to make computer/machines as intelligent as human beings in understanding language.
Now that you have a basic, and hopefully not too complicated definition of AI and NLP, you might ask why do we need to use NLP?
Today we are generating more data, often called big data, then we can humanly manage. Today around 80 % of total data is available in the raw form. Big Data comes from information stored in big organizations as well as enterprises. For example, for real estate professional, it includes location data, performance data, professional data like certifications and academic qualifications, contact data like social media, employment data like their previous position and so on. This huge amount of data can be analsed for patterns to better understand the information contained in data. NLP can solve big problems of the business world by using Big Data.
Natural language processing is a complex field and is the intersection of artificial intelligence, computational linguistics, and computer science. So how do we determine what the computer should write?
Firstly, that requires us to understand what we are writing about. Realtor Profiler understands which concepts it needs to convey through a teachable ontology, which includes the domain model for the relevant entities (including attributes and relationships) and the associated expressions for the domain model. For example, when writing a portal profile for a residential real estate professional, Realtor Profiler understands what a prospective lead wants to learn about the agent to activate that lead.
Secondly, Realtor Profiler is intent-driven, which means it starts with your goals: what do you want to achieve? Since Realtor Profiler knows the concepts pertinent to your domain, it can perform the types of analysis necessary to identify what is most interesting and important about your intent. In short, Realtor Profiler automatically runs all the computations necessary to arrive at the facts it needs to tell your story.
Thirdly, where is the data? Realtor Profiler needs to be directed to find data that is relevant to your intent, and then how to model this data against the ontology. To achieve this, Realtor Profiler asks data-specific questions to the real estate professional and then constructs language that conveys these facts in fluent, domain-specific language. Utilizing the same data source, Realtor Profiler can craft different narratives tailored to different audiences for LinkedIn, portal and agency websites. Realtor Profiler can also be customized by the tone of your business, creating fluid, natural language that fits an organization’s unique voice. Also, users can tell Realtor Profiler when to pursue certain types of information through the use of story conditions.
So in combination, these components enable Realtor Profiler’s system to scale to 12 million words a week without quality or consistency issues.
So is traditional manual ‘human’ writing doomed to extinction? The short answer is not now, but NLP is already powerful enough to service specific tasks and it is improving rapidly.
Traditional manual writing services are time-consuming. It requires scheduling, interviewing, editing and deadlines. Manual copy writing is also not scalable, so, for example, a small agency can hire a single person to write content about their key staff that is relatively consistent across multiple profiles. But large agencies cannot guarantee consistent branding across their organization’s thousands of employees because they would have to hire many individuals with different writing styles and levels of quality.
In contrast, NLP can generate content in seconds and also generate multiple variations so that a customer can choose the version they like most. The quality of the content is also better than most human writers. How? Because Realtor Profiler has invested about 15,000 hours into our content and that ‘knowledge’ is applied each and every time we generate narrative.
Traditional manual copywriting is also expensive, especially when the content needs to translate into another language. At Realtor Profiler, we can generate content for a fraction of the price of manual writing and price declines as volume increases.
The price difference becomes far greater when you considering translating into more than one language. At Realtor Profiler, we have been testing Google’s neural language translation technology. It enables us to translate accurately into Spanish, French, and Portuguese now. Within the year we will have added Mandarin, German, Korean, Hindi, Japanese, Russian, Turkish and Vietnamese. All of these will be available at the press of a button, instantly generated and highly accurate.
So in summary, AI and NLP are new to most people, but it is important to grasp a basic understanding of this new technology because it will change the way we communicate in the years to come.