Our founder and CEO, Peter Fitzsimmons, wrote a guest blog for Association Now’s newsletter update. See the original article here.
To be successful, all associations need strong member engagement strategies. One tactic is to use natural language processing and machine learning—two artificial intelligence technologies—to make member content feel more personalized and relevant.
Natural language processing (NLP) and machine learning are among the buzziest topics in technology and media today. There’s a lot of talk about using complex algorithms to produce magical results, like helping doctors to spot cancer, talking with your refrigerator, or having your Tesla drive you to work.
For many association executives, what these technologies can do at a practical level feels elusive, but there are real-world applications that can help associations succeed.
NLP and machine learning represent two branches of artificial intelligence that deal with the use of computer algorithms and technology to read plain text and unstructured content with human-like reasoning. Examples of plain text and unstructured content include articles, blogs, and white papers rendered or captured in a digital format. These technologies understand the context and meaning of huge volumes of data, including all of the rich media and information that your association shares with members 24/7.
But how can NLP and machine learning be incorporated into an association’s current operations, and why is it so critical to improving the member experience?
NLP and machine learning will drive the understanding of unstructured content created and delivered by your association and apply social media analysis, member engagement (active and passive participation), and even the analysis of third-party content that you may deliver from open public and private sources (in the intelligence community this is referred to as open-source intelligence).
Here’s how NLP and machine learning can help an association boost its engagement strategy with members.
The Netflix Problem
A lot has been written about the black-box algorithms used by Netflix, Google, Facebook, and other tech companies. Many people wonder how they know what you and other similar users are interested in watching, reading, or buying.
Here’s the real problem: Often, these companies don’t truly know who the person is on the other side of the screen. However, they’re still able to present meaningful suggestions about what that person might be interested in, using inferences and big data. But they can’t specifically understand each user’s interests, especially when there are multiple people using the same account.
Natural language processing can match subtle, nuanced, and specific content preferences and choices for members, precisely targeting them at an individual level.
For associations, NLP and machine learning can provide specific information that will help you understand each individual member’s preferences and interests and will directly factor into your future decision making. Developing a user profile for each member and analyzing how they interact and engage with the organization is essential.
NLP can organize and classify association content and media to match a specific member’s interest areas. The NLP engine can specifically target each member with highly relevant articles and information, as well as supplement that content with information coming from outside sources.
This technology requires much more than high-level content category tags. Semantic indexing in NLP can match subtle, nuanced, and specific content preferences and choices for members, precisely targeting them at an individual level. As members interact with the content, indicating whether they like or dislike the recommendations, NLP will use this feedback as a form of supervised machine learning. Each like or dislike provides direction on how a user’s interests are changing, and each recommendation gets better as time goes on.
Why is this critical? First, NLP and machine learning tools provide a richer and more personalized experience for members. Second, the tools inform staff about the content and media they might continue to invest in to provide high-quality experiences.
Better Search Results
The contextual search capabilities of NLP also make basic search functions much more effective in providing relevant results to members when they are searching for online content.
More than just keyword or Boolean search, contextual search enables members to express in their own natural language the topic or concept they are most interested in. Then, NLP engines can extrapolate these concepts using human-like interpretation and apply it to all of the association’s content, producing relevant results.
With the implementation of NLP and machine learning technology, associations will be able to enhance member engagement and experience, becoming more effective and meaningful platforms for long-term loyalty and affiliation. The results will energize your member engagement and reinforce your commitment to your members.