Categories
Blog

Which Search Engine Has the Best Results

I was recently wondering which of the popular web search engines provided the best results and decided to try to design an objective benchmark for evaluating them. My hypothesis was that Google would score the best followed by StartPage (Google aggregator) and then Bing and it’s aggregators.

Usually when evaluating search engine performance there are two methods I’ve seen used:

  • Have humans search for things and rate the results
  • Create a dataset of mappings between queries and “ideal” result URLs

The problem with having humans rate search results is that it is expensive and hard to replicate results. Creating a dataset of “correct” webpages to return for each query solves the repeatability of the experiment problem but is also expensive upfront and depends on the human creating the dataset’s subjective biases.

Instead of using either of those methods I decided to evaluate the search engines on the specific task of answering factual questions from humans asked in natural language. Each engine is scored by how many of its top 10 results contain the correct answer.

Although this approach is not very effective at evaluating the quality of a single query, I believe in aggregate over thousands of queries it should provide a reasonable estimation of how well each engine can answer the users questions.

To source the factoid questions, I use the Stanford Question Answering Dataset (SQuAD) which is a popular natural language dataset containing 100k factual questions and answers from Wikipedia collected by Mechanical Turk workers.

Here are some sample questions from the dataset:

Q: How did the black death make it to the Mediterranean and Europe?

A: merchant ships

Q: What is the largest city of Poland?

A: Warsaw

Q: In 1755 what fort did British capture?

A: Fort Beauséjour

Some of the questions in the dataset are also rather ambiguous such as the one below:

Q: What order did British make of French?

A: expulsion of the Acadian

This is because the dataset is designed to train question answering models that have access to the context that contains the answer. In the case of SQaUD each Q/A pair comes with the paragraph from Wikipedia that contains the answer.

However, I don’t believe this is a huge problem since most likely all search engines will perform poorly on those types of questions and no individual one will be put at a disadvantage.

Collecting data

To get the results from each search engine, I wrote a Python script that connects to Firefox via Selenium and performs searches just like regular users via the browser.

The first 10 results are extracted using CSS rules specific to each search engine and then those links are downloaded using the requests library. To check if a particular result is a “match” or not we simply perform an exact match search of the page source code for the correct answer (both normalized to lowercase).

Again this is not a perfect way of determining whether any single page really answers a query, but in aggregate it should provide a good estimate.

Some search engines are harder to scrape due to rate limiting. The most aggressive rate limiters were: Qwant, Yandex, and Gigablast. They often blocked me after just two queries (on a new IP) and thus there are fewer results available for those engines. Also, Cliqz, Lycos, Yahoo!, and YaCy were all added mid experiment, so they have fewer results too.

I scraped results for about 2 weeks and collected about 3k queries for most engines. Below is a graph of the number of queries that were scraped from each search engine.

Crunching the numbers

Now that the data is collected there are lots of ways to analyze it. For each query we have the number of matching documents, and for the latter half of queries also the list of result links saved.

The first thing I decided to do was see which search engine had the highest average number of matching documents.

Much to my surprise Google actually came in second to Ecosia. I was rather shocked with this since Ecosia’s gimmick is that they plant trees with the money from ads, not having Google beating search results.

Also surprising is the number of Bing aggregators (Ecosia, DuckDuckGo, Yahoo!) that all came in ahead of Bing itself. One reason may be that those engines each apply their own ranking on top of the results returned by Bing and some claim to also search other sources.

Below is a chart with the exact scores of each search engine.

Search EngineScoreCount
Ecosia2.820871778555523143
Google2.653978159126363205
DuckDuckGo2.583777012214223193
StartPage2.557232704402523180
Yahoo!2.512204424103742622
Bing2.48093753200
Qwant2.32365747460087689
Yandex1.926519337016571810
Gigablast1.51381215469613905
Cliqz1.397241379310342900
Lycos1.209626787582842867
YaCy0.8980503655564582462

To further understand why the Bing aggregators performed so well I wanted to check how much of their own ranking was being used. I computed the average Levenshtein distance between each two search engines, which is the minimum number of single result edits (insertions, deletions or substitutions) required to change one results page into the other.

Edit distance matrix of different search results

Of the three, Ecosia was the most different from pure Bing with an average edit distance of 8. DuckDuckGo was the second most different with edit distance of 7, followed by Yahoo! with a distance of 5.

Interestingly the edit distances of Ecosia, DuckDuckGo, and Yahoo! seem to correlate well with their overall rankings where Ecosia came in 1st, DuckDuckGo 3rd, and Yahoo! 5th. This would indicate that whatever modifications these engines have made to the default Bing ranking do indeed improve search result quality.

Closing thoughts

This was a pretty fun little experiment to do, and I am happy to see some different results from what I expected. I am making all the collected data and scripts available for anyone who wants to do their own analysis.

This study does not account for features besides search result quality such as instant answers, bangs, privacy, etc. and thus it doesn’t really show which search engine is “best” just which one provides the best results for factoid questions.

I plan to continue using DuckDuckGo as my primary search engine despite it coming in 3rd place. The results of the top 6 search engines are all pretty close, so I would expect the experience across them to be similar.

Categories
Blog

Why I quit using Google

So I was recently asked why I prefer to use free and open source software over more conventional and popular proprietary software and services.

A few years ago I was an avid Google user. I was deeply embedded in the Google ecosystem and used their products everywhere. I used Gmail for email, Google Calendar and Contacts for PIM, YouTube for entertainment, Google Newsstand for news, Android for mobile, and Chrome as my web browser.

I would upload all of my family photos to Google Photos and all of my personal documents to Google Drive (which were all in Google Docs format). I used Google Domains to register my domain names for websites where I would keep track of my users using Google Analytics and monetize them using Google AdSense.

I used Google Hangouts (one of Google’s previous messaging plays) to communicate with friends and family and Google Wallet (with debit card) to buy things online and in-store.

My home is covered with Google Homes (1 in my office, 1 in my bedroom, 1 in the main living area) which I would use to play music on my Google Play Music subscription and podcasts from Google Podcasts.

I have easily invested thousands of dollars into my Google account to buy movies, TV shows, apps, and Google hardware devices. This was truly the Google life.

Then one day, I received an email from Google that changed everything.

“Your account has been suspended”

Just the thing you want to wake up to in the morning. An email from Google saying that your account has been suspended due to a perceived Terms of Use violation. No prior warning. No appeals process. No number to call. Trying to sign in to your Google account yields an error and all of your connected devices are signed out. All of your Google data, your photos, emails, contacts, calendars, purchased movies and TV shows. All gone.

I nearly had a heart attack, until I saw that the Google account that had been suspended was in fact not my main personal Google account, but a throwaway Gmail account that I created years prior for a project. I hadn’t touched the other account since creation and forgot it existed. Apparently my personal Gmail was listed as the recovery address for the throwaway account and that’s why I received the termination email.

Although I was able to breathe a sigh of relief this time, the email was wake up call. I was forced to critically reevaluate my dependence on a single company for all the tech products and services in my life.

I found myself to be a frog in a heating pot of water and I made the decision that I was going to jump out.

Leaving Google

Today there are plenty of lists on the internet providing alternatives to Google services such as this and this. Although the “DeGoogle” movement was still in its infancy when I was making the move.

The first Google service I decided to drop was Gmail, the heart of my online identity. I migrated to Fastmail with my own domain in case I needed to move again (hint: glad I did, now I self host my email). Fastmail also provided calendar and contacts solutions so that took care of leaving Google Calendar and Contacts.

Here are some other alternatives that I moved to:

Migrating away from Google was not a fast or easy process. It took years to get where I am now and there are still several Google services that I depend on: YouTube and Google Home.

Eventually, my Google Home’s will grow old and become unsupported at which point hopefully the Mycroft devices have matured and become available for purchase. YouTube may never be replaced (although I do hope for projects like PeerTube to succeed) but I find the compromise of using only one or two Google services to be acceptable.

At this point losing my Google account due to a mistake in their machine learning would largely be inconsequential and my focus has shifted to leaving Amazon which I use for most of my shopping and cloud services.

The reason that I moved to mostly FOSS applications is that it seems to be the only software ecosystem where everything works seamlessly together and I don’t have to cede control to any single company. Alternatively I could have simply split my service usage up evenly across Google, Microsoft, Amazon, and Apple but I don’t feel that they would have worked as nicely together.

Overall I’m very happy with the open source ecosystem. I use Ubuntu with KDE on all of my computers and Android (no GApps) on my mobile phone. I’ve ordered the PinePhone “Brave Heart” and hope to one day be able to use it or one of its successors as a daily driver with Ubuntu Touch or Plasma Mobile.

I don’t want to give the impression that I exclusively use open source software either, I do use a number of proprietary apps including: Sublime Text, Typora, and Cloudron.