<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Mauricio P on Medium]]></title>
        <description><![CDATA[Stories by Mauricio P on Medium]]></description>
        <link>https://medium.com/@mauriciopt?source=rss-d1d4f1caf824------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*5yI0jpNHl-F_dy5B5Qi8Lw.jpeg</url>
            <title>Stories by Mauricio P on Medium</title>
            <link>https://medium.com/@mauriciopt?source=rss-d1d4f1caf824------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 21:05:46 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mauriciopt/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[NLP and Chatbots are not a happy couple, yet.]]></title>
            <link>https://medium.com/chat-bots/nlp-and-chatbots-are-not-a-happy-couple-yet-a6ca5986d7de?source=rss-d1d4f1caf824------2</link>
            <guid isPermaLink="false">https://medium.com/p/a6ca5986d7de</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[bots]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[nlp]]></category>
            <dc:creator><![CDATA[Mauricio P]]></dc:creator>
            <pubDate>Wed, 07 Sep 2016 23:18:28 GMT</pubDate>
            <atom:updated>2016-10-12T23:40:17.825Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FVxXv1jVJXpOyTmbPC-eqQ.jpeg" /></figure><p>I was at the grocery store, thinking about chatbots. Well, not really. I was fooling around with a new message app, and it was lousy with bots offering me gambling, webcam models, or a hook up. For most people, those things are annoying, but I always enjoy replying a couple times just to see how smart they are. These ones weren’t very smart.</p><p>The lines at the checkout were long, because there were only two cashiers working at the time. One wasn’t at her register, but instead filling the impulse-buy coolers with soda/pop. But the teenage kid ringing up orders fell behind, and a line built up. This didn’t bother me, because I went to the self-checkout aisle.</p><p>I watched the people in line as I bagged and paid for my groceries, the robotic voice guiding me through the process so I didn’t even have to think. They flipped through magazines, or diddled around on their smartphones. I got a ping from another bot. I didn’t even check it, just uninstalled the app, but it got me thinking.</p><p>The reason we have self-checkout at stores is because the tech-needed for it became more advanced. Now, people can choose to wait to interact with a real person or engage with a computer — who has a human’s voice — to complete the transaction. People seem to like it. Yet, when people encounter a customer service bot online, they don’t.</p><p>At Lexalytics, we have been doing natural language processing for a long time. We’re one of the leaders in the field. Our tech is out of this world. And we know a lot about how customers engage with companies online. We see it when monitoring social media or analyzing customer surveys. From providing voice of the customer examination to all aspects of managing the customer experience, we see where chatbots — like the self-checkout aisle — could become a great option for both keeping clients happy and focusing your human resources where they’re needed most.</p><p>Of course, if this was the 1960s or the 1970s, the self-checkout aisle would be impossible to achieve. Payments weren’t electronic, so you had to have an employee watching the money. The interface to the cash register was manual. Don’t let anyone fool you, retail clerks back then were skilled labor. Sometimes, in relation to chatbots, it feels like we just left that era.</p><p>The component pieces are there, but they just don’t quite work together in the most efficient ways. We know, thanks to our breakthroughs with Semantria and Salience, that we can help fix some of those problems. Yet, we can’t do it alone.</p><p>Chatbots have come a long way in recent years, and the market knows that eventually they will get where they need to be. Yet, it seems like every time there is an advancement, critics pounce and make the technology seem ridiculous. Recently it was Microsoft’s Tay AI. It was good, if a little repetitive, and it had a major weakness that internet trolls were able to exploit in less than a weekend.</p><p>For those of us who know about how technology develops and grows, we know these aren’t “defeats” or even all that “embarrassing.” No software was just created in its perfect form. Still, setbacks are to be expected, and Lexalytics has what it takes to make sure the mistake is not repeated.</p><p>To be fair, there are other natural language processing tools out there created by other companies that aim to improve how these chatbots work. Yet, also in fairness, these tools are incredibly weak. Some of this is the technical limitations of the products currently available. Some of them, without naming names, lack basic elements like proper entity extraction. However, there is also a problem with how these developers expect chatbot developers to use their tools.</p><p>We have the foundational technology that could be applied to chatbots, making them smarter and more intuitive. But we don’t build chatbots. Lexalytics is a sentiment analysis company at heart. We love AI as much as the next nerds, but we don’t know the best tool or platform to connect these worlds in a way that addresses not just what customers want, but what they don’t even yet know they need.</p><p>Others who develop NLP tools for use with chatbots is to provide developers with a tool that works a specific way, regardless of what they need. Essentially, they are saying “here’s our product, figure out how to make it work with your project.” We don’t agree with this approach. Rather than forcing chatbot developers to adapt their tech to our product, we want to help adapt our product to work with your tech.</p><p>So, this is my “open invitation” for bot creators to contact us, and maybe we can make this particular breakthrough together. You can try out our technology, run our tools through the paces, work with our online resources. We’re the leaders in NLP, but we don’t know best how to help you bot creators out there. At least, not yet.</p><p>Try out our product, see how it can work already with your chatbot project. However, if there are limitations or features that are missing from it that could make your chatbot work better, we want you to tell us. You know what you need, and it’s our job to ensure that we cater our services to address those needs. Using Salience or Semantria should solve the problems you are facing, not create a whole new set of obstacles to overcome. The only way we can pull this off is together.</p><p>Great discovery comes from collaboration. We know, because we have some damn fine teammates that make up the Lexalytics family. Right now, we’re living in a time of technological advancement that can only compare to how early humans figured out the practical applications of little inventions like the wheel or the discovery of fire.</p><p>This is a rare time for creators and big thinkers. Let’s not waste it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/725/1*xVXWVhL3QjUAZGByCnwf5A.gif" /><figcaption>Enjoyed the article? Click the ❤ below to recommend it to other interested readers!</figcaption></figure><figure><a href="https://docs.google.com/forms/d/e/1FAIpQLSfJDqeR7zbRBfQvx6mouQAhjK4lrCy3z9Z6Ppxln7f1xi12dg/viewform?c=0&amp;w=1"><img alt="" src="https://cdn-images-1.medium.com/max/255/1*rRjUc6JW61AP1UWsa-yItQ.png" /></a></figure><figure><a href="https://chatbotsmagazine.com/be-featured-in-front-of-thousands-of-people-interested-in-bots-e7040c4080df#.hswzhzpp9"><img alt="" src="https://cdn-images-1.medium.com/max/255/1*OGAA3ICxv3ds-RK2Sq4oxA.png" /></a></figure><figure><a href="https://chatbotsmagazine.com/want-to-receive-the-best-chatbot-related-content-in-your-medium-feed-20a9c11f11b6#.1t1yhwcyz"><img alt="" src="https://cdn-images-1.medium.com/max/255/1*1XecGp_cGZ1wGLAldRHLWQ.png" /></a></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a6ca5986d7de" width="1" height="1" alt=""><hr><p><a href="https://medium.com/chat-bots/nlp-and-chatbots-are-not-a-happy-couple-yet-a6ca5986d7de">NLP and Chatbots are not a happy couple, yet.</a> was originally published in <a href="https://medium.com/chat-bots">Chatbots Magazine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>