Unicorn NLP

Deep AI for Travel Reviews

Unicorn NLP

Deep AI for Travel Reviews

AI In-the-Box for Hotel Reviews - 124 precisely crafted Semantic Models

Products -> AI In-the-Box for Hotel Reviews

Unicorn AI-in-the-box for Hotel Reviews

AI In-the-Box for Hotel Reviews

  • Get the On-Premise version with 0 cents / text (Big Data compatible)
  • Trained and tested on hundreds of thousands of accommodation reviews
  • Trained and tested on your data (you get ready-to-use technology)
  • 124 Semantic models designed especially for hotel reviews
  • Capture 90% of information contained within hotel reviews
  • Proven state-of-the-art accuracy (precision=90-95%, recall=70-85%)

What is inside AI In-the-Box for Hotel Reviews

AI In-the-Box for Hotel Reviews consists of 124 Semantic Models. Each Semantic model was especially designed, built, tested, and re-tested on hundreds of thousands of accommodation reviews from 10 different sources. All presented Semantic Models work with an unparalleled precision of 90-95%.

Below, you will find a high-level view of all semantic models designed for hotel reviews.
In other words, what type of information do we garner from reviews.

Client loyalty

Will he come back? vs. not come back?
Will he recommend? vs. not recommend?
Overall good/satisfied. vs. Disappointed?

Hotel Opinions

How was the hotel? Was lobby clean? dirty? Was it nice? Smelly? Renovated? Outdated?
Was a visit satisfying? Is this hotel was a hidden gem?

Critical opinions / Alerts fot the owners

Bed bugs? Theft? Felt secure?
Hotel was not as advertised? Misleading info?
Dirty hotel? Service unpleasant/unprofessional? Didn't care?
Contact problem with Manager? Problem with refund?

Room&Bathroom Opinions

Room - Was it spacious? Was it clean? Was it comfy?
Was it outdated? renovated? Does it have amenities?
Bathroom - Was it clean? Soap provided? Poor shower pressure?
Kitchen amenities - is there a fridge, coffee maker, cups?

Opinions about Location

Opinion In General - is it a good location?
Close to: City centre? Public transportation? Restaurants? Shopping? Airport? Highway? Nature?
Neighbourhood: Is it a good neighbourhood? Is it safe?

Service Opinions

Staff - Are they friendly? professional?
Manager - friendly? professional? available?
Check-in - is it smooth, fast or slow?
Valet Service, Cleaning Service - good? friendly? professional?

Opinions about Accessibility

Handicap accessible or complains?
Elevator - available, functioning or broken?
Bus Shuttle - does it work? is it regular?
Parking, - is it ok? Is it big?
Valet Parking Service - available? friendly?

Addons Opinions

Wifi - is it free? Is it fast? Works in the room?
Pool - is it open? Is it working? Is it clean?
Gym - is it well-equipped? big? Or inadequate?
TV - is it new? Big? A lot of tv channels?


Problem with payments? with a credit card?
Problem with refund? problem with the bill?
Charged additional fees? Hidden fees?

Food Opinions

Breakfast: Varied and abundant? Or maybe small? Tasty? Free as promised? Did you have to pay extra?
Food: Is it adequate? Is it expensive? Or better to eat elsewhere?

If there will be need, we can expand this list.

What this technology can do (for business/hotel owners and for travelers)

1. AI-powered new analytical dashboards (for business owners)

1.1. Automatic Conversion of Reviews into Recommendations/Tips (for Business Owners)

First, you process all reviews of a given apartment and extract important information with high accuracy. Then, it is easy to compare these results with other apartments, and valuable patterns emerge.

Our AI deeply analyzed all 4423 of your reviews, compared with 450 000 reviews of similar businesses and extracted 14 Recommendations/Tips how you can improve your business:

  • Consider making bigger portions of the breakfast (7% complained in reviews - compared to 4.5% average) - read more and see reviews...
  • Repair the elevator (5% complained) and keep the pool open for a whole year (9% complained) - read more and see reviews...
  • Consider improving the speed of the WiFi (11% complained) and make it work in the room, not only in the lobby (5% complained) - read more...
  • Talk with your Manager so he could be more available for your clients (8% complained) - read more and see reviews...
  • Train your staff to be more friendly (12% complained) and more professional (4% complained) - read more and see reviews...
Extract more Recommendations from Reviews...

Compare your parameters with similar businesses

1.2. Human-like (Semantic) classification of reviews (for Business Owners)

Here, we focused on showing classified negative opinions critical to hotel owners and presented on a brand reputation dashboard.

By clicking on a specific semantic category (here: Noisy rooms) we get specific sentences/segments and "not the whole review" where people mentioned noisy rooms. We detect all the information that is correlated to noisy rooms (e.g. cleaner is banging around), not just with keywords (e.g. noisy, loud). This is the advantage if you use Human-like Analysis (Semantic Analysis) instead of keywords.

1.3. Human-like (Semantic) comparison of 2 hotels (for Business Owners)

Below is a simple comparison made only using 200 reviews and limited categories/areas. You have 124 Semantic Models at your disposal which can be used to make comparisons in general or in very specific areas.

Marriott, San Francisco (8.5 on booking.com)vs. Hilton, San Francisco (7.8 on booking.com)

Cleanliness comparison:

room clean 25 dirty 1
bathroom clean 1 dirty 10
cleaning service good 4 poor 4
bed dirty 1

Positive: 30, Negative: 16

room clean 27 dirty 14
bathroom clean 1 dirty 7
cleaning service good 5 poor 16
bed dirty 6

Positive: 33, Negative: 43

Service comparison:

staff helpful, friendly 94 unhelpful, unfriendly 9
check-in fast 8 slow 1

Positive: 102, Negative: 10

staff helpful, friendly 57 unhelpful, unfriendly 35
check-in fast 22 slow 8

Positive: 79, Negative: 43

Addons comparison (comparing only negative opinions):

wi-fi not free 20 problem 2 slow 1
parking not free 12 small 1

Negative: 37

wi-fi not free 9 problem 5 slow 1
parking not free 16

Negative: 31

Clients Opinion comparison:

opinion overall good, satisfied 6 will return 4 will recommend 2 overall disappointed 2 will not return 3

Positive: 12, Negative: 5

opinion overall good, satisfied 4 will return 5 will recommend 4 overall disappointed 14 will not return 7 not recommend 3

Positive: 13, Negative: 24

Marriott, San Francisco wins (Positive: 144, Negative: 68)
over Hilton, San Francisco (Positive: 105, Negative: 141).

On Booking.com Marriott also got a higher rate (8.5) compared to Hilton rate (7.5).

2. AI powered travel search & discovery & chatbots (for travelers)

2.1. Travel Hotel/Accommodation Search personalization (for Travelers)

Search based on your individual preferences matched with opinions of similar people to you and data from reviews

What is important to you? Personalize your search

Our AI remember these parameters while reading millions of reviews and find you the best possible place

  • Friendly, Professional Staff, frequent Cleaning Service
  • Big, Well-equipped Gym
  • Fast WiFi, everywhere
  • Free, big, tasty Breakfast
  • Handicap accessible & working Elevator
  • No hidden fees, No problem with Refunds
  • Nicely decorated place, clean, updated rooms
  • Clean, Well-equipped Bathroom, no shower problems
  • Fast Check-in, professional front-desk
  • more...
  • Recommended for (based on reviews)
    Solo, With children, Business trip, Romantic trip, Anniversary, Wedding, Quiet stay, Pet friendly
  • Close to (based on reviews)
    Restaurants, Local Atractions, Airport, Highway,
    Public Transport., Shopping, City center, Nature

Our AI deeply analyzed 3 million reviews from Berlin to match your personal needs:

2.2. Travel chatbot fueled by reviews - Automatic learning of information and users opinions from reviews and using it in conversation.

Meet the Alchemist Traveler. He has been everywhere and he has read all reviews in the world (and he remembers them all).
He is patient and he can spare a moment to plan your perfect vacation.

Example 1:

Where are you going?
Great, please describe what it is important to You or select by tapping on the screen
working free wifi, working pool, pet friendly and no smelly rooms
Good choice! I have just analyzed 3 millions (wow! right?) reviews against your needs and I found 17 places.

Example 2:

Where are you going?
What is important to You?
I don’t like dogs, children, need elevator
Got it! Something more?
no hidden fees and deposits
I have just analyzed 3 millions (wow! right?) reviews against your needs and I found 12 places.

Example 3:

You selected Hotel A in Barcelona, do you have some questions about it?
oh yes, very important, is it handicap friendly for real?
Yes, 24 guests stated in reviews that it is handicap friendly, do you want to see specific sentences from reviews of that guests?
yes please
“i was satisfied that elevator was working properly because I was with my eldery aunt”, “staff of the hotel take very good care of my old papa”, bathroom was handicap accessible!”
great, book it!

2.3. Human-like (Semantic) summary of reviews (for Travelers)

Instead of reading thousands of reviews individually, you can see a semantic summary and get a feeling for what is contained within them in 20 seconds. You can click for specific information and see extracted sentences from reviews.


hoteloutdated 23 smelly 22 clean 15 not as advertised 4
opinionoverall disappointed 34 will not return 22 will recommend 9


roomclean 34 spacious 22 well equiped 15 sth broken 4
bathroomwell equipped 23 shower problem 19 clean 15 outdated 13 small 8


staffhelpful, friendly 23 unhelpful, unfriendly 15 check-in slow 12
restaurant servicepoor, slow 8
managerunhelpful, ineffective 8
valet servicebad, slow 23


breakfastrich, plenty options 13 not free as promised 9 great 6
pricefood overpriced 7


hoteloverpriced 23 money waste 15 adequate price 4
paymentbill problem 8 additional fees 5 refound problem 3


close tohighway 7 city center 5 public transportation 4
neighborhood not safe 14


wififree 11 problem 9 slow 7
tvbad, poor channels 13
parkingsmall 19 not free 17
poolclosed 8
handicapnot accessible 9


deal breakerssth stolen 4bugs 2

The automatic summary made from reviews - after processing by AI In-the-Box for Hotel Reviews (no additional NLP/NLU/ML or post-processing needed).

The first no-block title (Hotel, Price etc.) is a Category from our API, blue blocks (Bathroom, Payment etc.) are Aspects, green and red blocks (outdated, well equipped etc.) is the Feature and Polarity (Feature is text, Polarity is color).

See the live demo of semantic summary made from hotel reviews (powered by our technology).

3. Other solutions in Travel

3.1. Semantic Distribution and TagCloud out of 1000+ reviews (for travelers&business owners)

All displayed results make sense. By clicking on specific information, you can display actual sentences (not whole reviews).

Our AI deeply analyzed 4765 Reviews and organized them for you
(so you do not have to read them all)

People were the most satisfied with:
Rich Breakfast
Clean Room
Hotel Renovated
Well-Equiped Room
Fast Check-in
People complained about:
Pool Closed
Problem with Refund
Problem with WiFi
show me more...

3.2. Human-Like Client Loyalty Analysis (1-dimension deeper than Sentiment Analysis).

By understanding a review better, you are able to provide deeper and easier-to-understand analysis for your clients.

1. Sentiment Analysis (positive vs. negative, Black&White approach)

Positive vs. Negative


2. Human-like Analysis (semantic, COLOR approach)

Overall good/satisfied vs. Overall dissapointed


Will come back vs. Will not come back


Will recommend vs. Will not recommend


Language has color. Do not reduce it to black & white.

3.3. AI-powered Automatic Review Report Generation (for travelers&business owners)

Information extracted by 124 Semantic Models is precise enough to create automatic text summary.

Natural Language Generation Summary from Reviews could look like this:

Our AI deeply analyzed all your reviews for the past 3 months, combined them with reviews of similar businesses and created this summary:

Most of people (88%) was satisfied with their stay, a lot of people (47 mentions) wrote that they would come back and will recommend this apartment (24 m.).

Reviewers state that rooms in this apartment are clean (34 mentions) and spacious (22 m.) and well-equipped (15 m.). Unfortunately, reviewers also stated that the apartment was outdated (23 m.) and smelly (22 m.). The majority found the staff helpful and friendly (23 m.) but some stated that they were unprofessional (19 m.) and the check-in process (12 m.) was slow.

There were also some issues with the Pool (23m.) and the Gym (15m.)
Generally, people say it is a good value for money (42m.).

Semantic distribution summary out of text:
Overall good/satisfied
Would come back
47 mentions
Will Recommend
24 mentions
Clean Rooms
34 mentions
Spacious Rooms
22 mentions
Well-equipped Rooms
15 mentions
Outdated Hotel
23 mentions
Smelly Rooms
22 mentions
Friendly Staff
23 mentions
Unprofessional Staff
19 mentions
Slow Check-in
12 mentions
Issues with Pool
23 mentions
Issues with Gym
15 mentions
Good value for money
42 mentions

Want to talk more about these solutions?

Contact Us

How it works on a single review

Example hotel review:

The hotel was clean and renovated, service was friendly too. But that's it. The bathroom was dirty, Shower got poor pressure. I think there were fleas in our bed The pool was closed during our stay and staff could not tell us where it will be open and does not seem to care. The breakfast was a joke, the portions was small and we were told that we could pay to get more, so it was not free as promised. Wanted to get my money back but they refused, not coming back.

Simplified output after processing by AI In-the-Box for Hotel Reviews:

1. The hotel was clean and renovated, service was friendly too hotel clean +1
staff helpful, friendly +2
hotel renovated +1
2. But thats it
3. The bathroom was dirty, Shower got poor pressure bathroom shower problem -2
bathroom dirty -2
4. I think there were fleas in our bed deal breaker bugs -2
5. The pool was closed during our stay and staff couldn't tell us where it will be open and does not seem to care pool closed -2
staff unfriendly, unhelpful -2
6. The breakfast was a joke, the portions was small and we were told that we could pay to get more, so it wasn't free as promised breakfast poor, limited -1
breakfast not free as promised -2
breakfast bad -1
7. Wanted to get my money back but they refused, not coming back opinion will not return -2
payment refound problem -1
Live Demo and Comparison with other technologies

How we differ?

We specialize in creating dedicated Natural Language Understanding engine for specific reviews. We focused in travel and prepared a set of products to process many different travel reviews (e.g. hotel, hostel, apartment, restaurant, airline reviews).

We stand by the statement that we detect more information than Watson or than any other Deep Learning solution and we are more accurate than Google in Travel Domain. We are also much more cost-effective. And we allow the processing of as much data as you want with no additional cost (0 cents / text).

This is possible because we created the next-generaion process of developing NLU systems in which we can build Semantic Models 10 x faster. It is designed from the ground up to process reviews, not proper english texts. It is the result of 12 years of experience in NLP/NLU and 8 years in processing reviews, and more than 50 big implementations (2007-2019) of NLU systems to process reviews and other user generated content.

We are passionate and proud of our technology and aim to provide the optimal technology for travel reviews. We will be happy to help you implement your brilliant ideas and discover what is possible.

1. Sentiment Analysis & Concept/Keyword/Topic Analysis - Current tech available

"Breakfast was tasty" => Breakfast: positive
"Breakfast was huge" => Breakfast: positive
"Breakfast was not free" => Breakfast: negative
"Breakfast was very tasty but really small" => Breakfast: neutral
"Breakfast was huge but you had to pay extra" => Breakfast: neutral

2. Human-like Analysis (AI In-the-Box) - Unicorn NLP

"Breakfast was tasty" => Breakfast: tasty
"Breakfast was huge" => Breakfast: big portions
"Breakfast was not free" => Breakfast: free
"Breakfast was very tasty but really small" => Breakfast: tastysmall portions
"Breakfast was huge but you had to pay extra" => Breakfast: big portionsnot free

Language has colors. Do not reduce it to Black & White.

Technological differences (compared to other NLP/NLU solutions):

  • High Resolution/depth of output, richness of output, what different type of information we capture out of reviews - state-of-the-art: 124 detailed Semantic models
  • Very high Fact Coverage (state-of-the-art: 90%) - percentage of pertinent information we extract from reviews
  • Very high, Human-like Accuracy - state-of-the-art F1: precision=90-95%, recall=70-85%
  • Detailed and easy-to-use Semantic models (e.g. comfy room, spacious room, tasty breakfast, will come back) instead of statistical models (e.g. room - confidence: 0.84623) or grammatic models (e.g. NS, VP, ADV)
  • Very fast! We are much much faster than any deep learning or any other solution. Highly optimized technology was very important for us from the beginning. To show you how much, take this example: On a Amazon AWS medium-cpu instance we process up to 1 million reviews a day.

Other differences:

  • You can get the on-premise version with no maintanance fees and 0 cents/text
  • No linguistic or machine learning expertise required to implement it into your product/system! Output is simple enough to be used by anyone and even simple enough to be displayed to the user directly (no post-processing needed, no configuration required, most solutions can be implemented in days)
  • All data extracted by our tech makes sense - we do not provide hard to understand keywords with confidence and some of them do not make sense. Each Semantic Model is actionable data (tasty breakfast, wifi does not work, staff was unfriendly, elevator not working, etc.). Each Semantic Model was built and tested on hundreds of thousands of hotel reviews. Each Semantic model consist of between a hundred and a thousand of ways to express concrete situation. That is why you do not have to configure it to your data.

More about Unicorn NLP Technology

What is a Semantic Model

The semantic model is the newest approach to an information extraction models where boundaries of those models are not determined by the structure of a language (or grammar, or used words), but the type of information you are detecting. It does not matter how the user writes an opinion about a specific object, e.g. “breakfast was a joke”, “Muffin and cold coffee is not a breakfast”, “Breakfast wasn’t included as written”, our semantic model still detects it and provides structured-data. With "breakfast" example, it is relatively easy to provide a shallow analysis (e.g. Sentiment Analysis) with the current NLP techniques. If you are looking for more sophisticated information (e.g. is it safe?, is it handicap friendly?, will someone come back?, Is it sustainable? Is it pet-friendly?) the situation gets more complicated. Current techniques provide a part of this information and their accuracy relies on the keywords used in reviews. Users are very creative in the reviews and they are not using a keyword approach. Let’s take one example. If we want to answer the question “is it safe?”, you need to capture a lot of information from reviews without common keywords safe/dangerous. In practice they write this in tens or hundreds of ways: e.g. “there was a lot of drunken people outside”, “I did not feel good because of the neighbors”, “someone yelled in the night and the front desk lady did nothing”. We believe, that if the human can understand what is written in the review, then technology should detect it as well. That is why we call them Semantic Models, not information extraction models.

Where are we making these Semantic Models

Our UnicornNLP Semantic Environment is a place where we design and precisely craft all the semantic models. It is a perfect combination of human and computer intelligence. It is the secret sauce of our technology and it allows us to deliver the best technology in the world for reviews. Statistical algorithms provide the data based on the processing of hundreds of thousands of reviews, but the humans are making all the crucial decisions. Given the pace of AI development, this is significant advantage in information extraction (in NLP/NLU) and it will be appropriate for the next decade. Think about our environment as a way of creating tens of thousands of very precise, hand-crafted semantic rules but with an AI as an assistant that is generating new ideas. These ideas are provided by an AI assistant based on constant simulations on real data and statistical tools.

We designed a semantic programming language (QL4Reviews) to extract information from reviews. With this programming language, you build semantic models 10 times faster than in other current technologies. Every Semantic Model is programmed in QL4Reviews and consist of 50-2000 lines of semantic code. Heavily optimized engine allow to process through 124 Semantic Models on 1 low-cpu Amazon AWS instance - 1 million reviews a day. Speed of the core engine was very important for us from the beginning.

How to process non-English Reviews

Our technology is designed from the ground up to process reviews, not proper-grammar texts. It can process also machine-translated reviews/texts (via Google Translate or Microsoft Translate). This approach provides almost as good results as processing English reviews but we do not have to rewrite deep semantic models into other languages. It is much easier to maintain and scale (because there is only 1 engine).

Read more & See demo: Processing machine-translated non-English Travel Reviews

Are we another NLP API for Travel?

This is not just another NLP API/Tool. We redesigned every layer of technology and we wanted to address the biggest challenge in the NLP/NLU industry. We wanted to change the way you use and think about Natural Language Understanding API. In our opinion, if you are using NLP API, you should not get your hands dirty and spend inordinate time thinking how to best implement this API into your application. We wanted to hide all the NLP complexity inside the system and provide you ready-to-use data. We believe that NLP API should be simple and easy to use without linguistic, machine learning or other expertise. You do not need to spend weeks to improve the accuracy of "domain-indepentent" NLP API, to adjust to Your domain or to make decisions about complex output parameters. Getting output parameters like confidency (e.g. 0.7852) or polarity (e.g. 0.4352) or keywords detected (e.g. friendly, 0.6786; dissapointed, -0.7564) gives every developer problems and frankly they are not standard or predictable. They are rather a consequence of used technology inside, and putting these parameters for the output in our opinion is the way to leave configuration and responsibilty for decision-making to You. We believe in human-like output where a human can understand what this API detected. We believe that we have developed excellent API for specific domains, so you can get tailored results and focus on your business. We position ourselfes as a read-to-use technology, not a standard API that is a tool that you need to configure and train to your data.

What's next

If you want to buy a Product, receive more info about the On-Premise version (AI In-the-Box) and Plans&Pricing
or want to discuss what is possible

Contact us

If you want to test it or see how it works on your data, send us a dataset
(we do not collect your data)

Test on your data

You can also start with the SaaS version - pay per text (hosted via rapidAPI & Amazon AWS)
before you get the On-Premise version with 0 cents / text!

Link to RapidAPI