The Technology Stack that Runs Low Cost Hotels

We believe travel planning should be much easier. This is why we continuously improve the SmartCost® algorithm to explore and analyze the gigantic mountains of data that our automatic web crawlers collect. To do so we run various methods of big data analysis and Artificial Intelligence price predictions.

LowCostHotels.com's technology evolved from a research project led by computer science engineers with expertise in the travel industry vertical. We've spent several years developing and refining our state-of-the-art predictive, statistical and data exploration algorithms with the objective of providing consumers with unprecedented, actionable information to help decide which hotel provides best value-for-money and when is the best time to book.

Quality Signals Collection

Our web spiders collect hotel signals from many data sources. These sources feed our hotels ranking algorithm.

  • We collect raw information about individual hotels such as their location, list of amenities, images from our booking partners and/or the hotels themselves, etc..
  • We collect 2nd degree real-world parameters such as the hotel renovation date, management and chains structure, closest metro, neighborhood popularity and more.
  • We process millions of travel reviews that are gathered from all over the internet. Some reviews are collected popular sites such as TripAdvisor, Expedia, Booking while others from niche sites. We also monitor business rating sites like Yelp and Foursquare and also process individual travel bloggers reviews using NLP (natural language processing) methods.
  • We process social network signals from Facebook, Twitter, Instagram. This allows us to be on the edge with the latest changes.
  • In addition to the international sites mentioned above, we also read data from local sites in different languages (Ctrip, Yandex etc). We then translate this data and analyze it using NLP to mine relevant insights.
  • Last but not least, we analyze the hotel availability over time to calculate hotel popularity scores.

We already have millions of data points but we are constantly improving our technology to ensure you have one place to get the most trusted information you need.

Prices Data Collection

Each day LowCostHotels.com spiders query and collect millions of hotels price from many inventory sources. These price comparisons are run against online and offline sources and top GDSs.

The itineraries we query extend up to a 180-day period in advance, encompassing different trip lengths that span over 150,000 international destinations.

For each hotel, destination, trip length and guests combination pairs we compute the predicted prices trend. This is equivalent to performing thousands of searches on a typical travel search site and running millions of data processing formulas.

Data Aggregation and Analysis

The huge volume of hotel data we process every day is aggregated and transformed using a variety of statistical measures. These statistical measures allow us to intelligently filter the hotel data to reveal where the best prices and deals reside. We also calculate special features from our historical data that are predictive in nature. This data is used to train our predictive models which ultimately power the recommendations on our website.

New AI technologies along with the latest cloud compute capabilities operates the backend of the process-consuming SmartCost® algorithm. This data forms the foundations of our hotels price recommendations, predictions, deals, and other flexible travel features.

Modeling and prediction

Our Research and Development team is comprised of experts in the fields of statistics, data mining and BI analysis. We have developed algorithms that can identify patterns and conditions from our history of accumulated the data associated with significant price changes. These patterns, learnt from historical data and nearby hotels, are represented and stored in predictive models. Once trained, we use these models to predict the future fares in conjunction with current market conditions.

Measuring the Price Predictor Performance - Simulation Technology

In order to test our predictive accuracy we have developed robot-like applications that simulate customers booking hotel rooms based upon prices from real data we process and store every day (sometimes referred to as “back testing”). We build sequences of models for each day we have data and simulate a distribution of travellers with different economic constraints booking rooms, similar to those found in the real world. Each simulated customer receives a recommendation for the day and market they are shopping from the predictive models. We tally up the results of the outcomes of recommendations made to thousands of simulated customers. This data tells us how accurate we are and how much money was saved or lost by our simulated customers. Although the simulations do not mimic real life exactly, they provide a close enough approximation for us to understand the efficacy of our technology.

Visualizations

Given the multi-dimensional complex nature of the data we process, as well as the derived data sets we create, we rely heavily on data visualizations to help us understand rate trends and patterns. Being able to visualize time-series information in many dimensions can be a challenge with traditional visualization schemes. Oftentimes creative and unconventional forms of data images have provided the most potent snapshot of phenomena we are interested in researching. Further, data animations have been a useful tool for studying variables that change over time such as price.

Patent Pending Algorithms

LowCostHotels.com's prediction and recommendation technology and its proprietary features, functionality, and business processes are in the process of being patented in the U.S. and multiple EU markets.