What do people write about in customer product reviews?

Have you ever been in the situation that you wanted to buy something on amazon but you could not decide between product A and product B. Both products have a few hundred product reviews and you would like to have a summary of what all the customer reviews are saying?

Now there is a new web app for exactly that type of problem. IReadLess.com.
It reads customer reviews and extracts the product features the people are talking about and also if they are writing positive or negative about it.

IReadLess under the hood performs something which is called Sentiment Analysis on customer reviews. The term describes various techniques to recognize opinions and emotions in text. Normally this is used to classify a piece of text into the two categories positive or negative. IReadLess takes this one step further and also tries to extract the product features people are writing about and recognize if it is positive or negative.

So what can I use it for?

I think you can see it as kind of a shopping advisor. It gives an additional base to make a decision for or against a product. It is kind of an additional view on a product by crawling through all the customer reviews and extracts and summarizes what is written there. It saves you from doing that yourself and it is this high-level summarized view on a product which you do not have when you just check the 5-star ratings alone.

How does it work?

It is pretty simple. Drag the bookmarklet on their page into the bookmark-bar of your browser, view a product on amazon and click the bookmarklet which takes you to the ireadless result page for that product. If the product you were viewing has customer reviews then you will see two top5-lists of best and worst-rated product features.
It also displays a circle-chart showing the number of positive and negative ratings in total. This is useful for the first instant to see if the majority of mentions is positive or negative.

You then can click on each feature (like the sound of this headphone) and see the parts of the reviews it was mentioned in. In each sentence it was mentioned in you see the feature highlighted along with the words describing it as positive or negative. In some cases it is not 100% correct, especially with things like sarcasm which is a well known and very difficult problem in Sentiment Analysis and natural language processing (NLP) in general like this article describes.


If you are buying on Amazon a lot then this new tool may be interesting for you.

Disclaimer: I am one of the people behind IReadLess, but wanted to write here from a more personal perspective. Hope you like it anyway. Your feedback is always appreciated 🙂

Dieser Beitrag wurde unter Software-Development abgelegt und mit , , , verschlagwortet. Setze ein Lesezeichen auf den Permalink.