The Fight Simple Things to Make You Happier At Home

Prior to getting in place every day, establish the aiming for that day. With All the Craft regarding Peace, the particular Dali Lama states “Daily, feel whenever you wake: today Now i’m fortunate to be well, I have a precious individual existence, I am not able to use up it.” Wow. Nice to read a shrewd man. I tend to awake at the top of a stronger splanchnic problem saying, “Particular attention world: Turn out to be skeptical because of everyone prior to coffee. Come to be pretty too embarrassed!” Arranging an everyday set an extraordinary enormous difference. Your daily intention might be such as “wind up being productive” and “have fun with today’s mouth watering moments” or it can be something a lot more specific love “suggest give thanks to you the spouse and children today.” Though it should read this article not get a further “in order to do” merchandise within your list.

Read more

Machine Learning in Travel Industry

dlc. Travel Tech

Unless you’ve been living under a rock, ignoring every big tech advance in the past few years, you must have heard of machine learning!
For example, the recommendations made by Netflix and Amazon, the personal
assistants competing in our iOS and Androids devices, and  the futuristic self-driving cars are all forms of machine learning. Basically, machine learning is empowering the current artificial intelligence revolution. But what is it exactly? And how does the travel industry use it?

What is machine learning?

Machine learning is an application of Artificial intelligence (AI) which provides systems the ability to automatically learn and improve from experience, without being explicitly programmed.

Strange right? The co occurrence of “Machine” and “learn and improve from experience” in the same sentence, as if we’re talking about a human being , we kinda are, after all the goal of artificial intelligence is to mimic the human intelligence.

So, how does the machine learn after all?

This question is not yet easy to answer here, but let’s start by breaking down the big problem which is machine learning to different smaller set of main sub problems, the first one is called Supervised learning, the other one is Un-Supervised learning (there are other types but we’ll focus on those two to keep it simple).

Supervised machine learning, imagine you have a booking website, you’ll find yourself with a huge data about travel related reviews on hotels, places to visit etc. and you want to know the general sentiment on one of the hotels that you host on your website (maybe to list the hotels with the most positive feedback at the top of your search results).

Our supervised learning models can solve such task, these algorithms will look at as much as it can see from your customer positive and negative reviews, learning in the way the patterns that make a user review positive, this pattern may be finding the word “very” and the word “good” in the same phrase means that the sentence is positive, and it learns as well the pattern that makes a user review negative.

That’s why it’s called supervised, because our learning algorithms look at the data in a supervised way, they already have the answer (each phrase is positive or negative) and they try and learn the pattern that generates that answer, then our learning models are ready to be set free and look at new unseen hotel reviews and predict whether these reviews are positive or negative.


Un-supervised learning, unlike the previous type of learning, this one doesn’t use known outcomes to guide the learning. These algorithms can find patterns and relationships automatically in our data.

Imagine you want to send travel packages offers for your customers in a specific season maybe during the summer, but different people will react differently to the same package, unsupervised algorithms may come handy in these types of problems, where you want to cluster your customers into different customer segments and you don’t have any outcomes to guide the learning, then send each customer segment the package that they will probably love.


Machine learning at dlc. Travel Tech

Here at dlc. We’re integrating machine learning into our core apps, here’s a few examples:

a. Leads ranking – in Tangram CRM, you can classify new customers as hot or cold lead, in other words probability of whether the customer is gonna confirm a reservation or not.

b. Time Series Forecasting – Given historical data, we can provide predicted values to travel agencies that would forecast future revenue or total number of customers over time.

c. Recommender Systems – Recommend packages to customers, based on the information available about user, packages and user preferences.

d. Chatbots – Automated chat machines that can understand typed up words and sentences and communicate with customers to provide sales services.

If you’re interested to know more about the wonders of AI, check out this previous blog out.