Why some recommendations fail: recommendation engines and their challenges

We’re excited to bring Transform 2022 back in person on July 19 and pretty much July 20-28. Join AI and data leaders for insightful conversations and exciting networking opportunities. Register today!

Using algorithms to make purchase suggestions is big business. Netflix reported that its recommendation engine contributes to this $1 billion every year to the max. However, sometimes the suggestions are far off.

Take, for example, an ad I received to apply for a job as a van driver. I’ve never been a professional driver, I don’t even like driving and I’ve never owned a van. Obviously this recommendation engine knows nothing about me.

There are several ways in which recommendation algorithms can reach wrong conclusions. Here are just a few examples for each type of recommendation engine.

1. Collaborative Filtering

This filtering method is based on collecting and analyzing information about user preferences. The assumption is that if two users have one common interest, they also have other common interests, so product recommendations for both match. The advantage of this type of analysis is that the algorithm doesn’t need to use inferences from deep learning to understand the item being recommended, it just needs to identify users with similar interests.

However, a disadvantage of collaborative filtering is that it requires a large dataset of active users who have reviewed or purchased a product in order to make accurate predictions. If you have low user activity, it is much more difficult to generate good quality recommendations. The number of items sold on major e-commerce sites is extremely large. Therefore, even the most popular items can have very few ratings. This is considered the long tail or data scarcity problem.

There is also no way to process new items that have not been previously reviewed.

In addition, there are millions of users and products in many of the environments where these systems make recommendations. Thus, a large amount of computing power is often required to make the required calculations, meaning many companies are forced to limit the amount of data their models capture, which can negatively affect accuracy.

2. Content based filtering

Content-based filtering methods use keywords that describe an item to make a match between recommendations and people. For example, when recommending jobs, keywords from the job description can be matched with the keywords in the user’s resume.

The main disadvantage of this model is that it can only make recommendations based on the user’s existing characteristics. It also requires text analysis, which can introduce errors when the algorithm needs to identify keywords that are written differently; for example: instructor, trainer, teacher or supervisor.

This type of recommendation engine is also challenged when the solution is multilingual and requires translating and comparing words and phrases in different languages.

3. Hybrid Recommendation Engines

Hybrid recommendation systems use collaborative filtering and product-based filtering in tandem to recommend a wider range of products to customers with more accurate precision.

Hybrid recommendation systems can generate predictions individually and then combine them, or the capabilities of collaborative methods can be added to a content-based approach (and vice versa). In addition, many hybrid recommendation engines include demographics analytics and knowledge-based algorithms that draw inferences about user needs and preferences based on deep learning.

But even if hybrid recommendation engines can improve accuracy, they can suffer from longer compute times. The importance of speed differs per application. Movie and e-commerce recommendation systems, for example, may learn at a slower pace, while an application that recommends who to follow on Twitter will change regularly, forcing a recommendation engine to make near-real-time predictions based on new data.

In addition, personal interests have different levels of time sensitivity. For example, individual sports such as running or swimming are for the long term, while following sporting events such as championships for favorite professional teams can change constantly. Recommendations based on real-time matches should be updated more often.

Improving accuracy for all types of recommendation engines

In all cases, to be more reliable, recommendations must be varied, adapt quickly to new trends, and have the ability to scale quickly to handle more data. One way for developers to improve the accuracy of their recommendation engines is to use out-of-the-box, pre-trained models and invest in MLops tools that can help speed up the model production process and update models regularly. check to check for drift.

Personally, I’m always happy to see recommendations for restaurants, bars, books and music performances. Even if the predictions are far off, I can be convinced to try new things. But by using more complex models pre-trained with more data, I’m less likely to be asked to apply for a job as a van driver.

Michael Galarnyk is an AI evangelist at

DataDecision makers

Welcome to the VentureBeat Community!

DataDecisionMakers is where experts, including the technical people who do data work, can share data-related insights and innovation.

If you want to read about the very latest ideas and up-to-date information, best practices and the future of data and data technology, join us at DataDecisionMakers.

You might even consider contributing an article yourself!

Read more from DataDecisionMakers

Shreya Christina
Shreya has been with for 3 years, writing copy for client websites, blog posts, EDMs and other mediums to engage readers and encourage action. By collaborating with clients, our SEO manager and the wider team, Shreya seeks to understand an audience before creating memorable, persuasive copy.

More from author

Related posts


Latest posts

Quality vs. Quantity: The Trade-off in Data Annotation Without the Right Tools

A bug tracking tool or issue tracker, such as BugHerd, is a specialized bug tracking system designed to record and track website or software...

Data Security in the Digital Age: Best Practices for Backing Up Essential Information

In the age where digitalization has permeated everything, the sheer importance of data security cannot be overemphasized. Whether it may be personal pictures and...

10 Tips To Work From Home For Less

Over the past few years, the idea of remote work has seen a significant surge in popularity, providing employees with the valuable advantages of...