An important subset issentiment data— information on how people perceive the given product, event, idea, etc. The fundamental categories here are“perceive positively”and“perceive negatively”.
Until recently, sentiment data wasn’tquantifiable: It was impossible to measure people’s sentiments precisely. With the advent of natural language processing and machine learning, however, this task has finally become attainable.
In this article, we’ll explore how you can utilize sentiment analysis and web scraping to make better financial decisions.
Overview of sentiment analysis
Even the best industry professionals cannot keep up with all the latest news, reports, updates, and rumors. This data often drives the decision to, say, buy or sell the given company’s stock. Here’s a typical example:
Amid growing concerns about COVID-19, the government ofCountry Xdecides to use video conferencing instead of holding in-person meetings.
Video Conferencing Software Yis one of the most popular video conferencing solutions on the market, so the markets are expectingSoftware Yto acquire a plethora of new users.
Software Y’s rise in popularity is reflected in its stock price.
The scenario above borrows heavily fromZoom’s recent success, which can be illustrated by the following chart:
To a certain degree, the process of analyzing this data — news, reports, updates, and rumors — can be automated
. Upon noticing a headline like“Coronavirus: Zoom Video to hire 500 new software engineers as usage surges”, this software would act according to the guidelines we provided (e.g. buy Zoom stock.)
Tesla’s stock jumped 2.5% after Tencent said it amassed a 5% stake in the electric car maker. Ocwen jumped 12% premarket after disclosing it reached a deal with New York regulators that will end third-party monitoring of its business within the next three weeks. In addition, restrictions on buying mortgage-servicing rights may get eased. Cara Therapeutics’s shares surged 16% premarket, after the biotech company reported positive results in a trial of a treatment for uremic pruritus.
Another great example is the recenttweetof Elon Musk: “Tesla stock price is too high imo”.
This has decreased Tesla’s stock price. Notice the dip on the 1st of May:
The system that makes sentiment analysis possible is callednatural language processing(or NLP for short.) As their name suggests, NLP algorithms are designed to analyze the meaning behind texts innatural(i.e. human-made: English or Chinese) languages.
Although building and implementing an NLP system takes a lot of resources, the benefits make this endeavor worthwhile:’
The algorithm boasts superiorreaction time: it executes commands in mere milliseconds and works 24/7.
It also offers scalability: Its“expertise”can be applied to — given enough computing resources — every source of financial data.
How does sentiment analysis work?
Every text has a certain attitude, either positive, negative, or neutral. Sentiment analysis aims to determine the attitude of the given text (in most cases, of individual phrases and sentences) via splitting it into individual words (calledtokens), determining their attitude, and then determining the overall attitude of the target text.
This principle may seem confusing, so let’s play around with this technology ourselves.
Python programming language has an NLP-focused library called NLTK (Natural Language Toolkit).This websitefeatures an interactive implementation of NLTK’s sentiment analysis algorithm. Try inputting different sentences to see how the algorithm perceives them.
Let’s test the following sentences:
“This project is a great tool for processing raw data.”The algorithm determines that this text ispositive.
“This project will change the tech landscape.”The algorithm determines that this text isneutral.
“This project failed to live up to its potential.”The algorithm determines that this text isnegative.
Shortcomings of sentiment analysis algorithms
Previously, we used sentences with rather straightforward meanings in the interactive prompt: Words like“great”and“fail”usually mark the entire context. What about something more complex? Let’s try it out.
Let’s take this phrase as an example:“The automobile industry has seen better days.”The algorithm determines that this text isneutral.
These examples show that traditional NLP algorithms have a hard time parsing implicit meanings:
Enhancing sentiment analysis with machine learning
This is wheremachine learningcomes to rescue: We can train an ML algorithm on countless examples to make it“understand”the text’s context. Here’s a blueprint for such a project:
FBS CopyTrade Became the Best Application for Copy Trading in 2020Go to article >>
Collect a dataset that focuses on financial sentiment texts.
Mark up each text’s sentiment.
Build a sentiment analysis model that is optimized for“financial language”.
The basis for a machine learning algorithm lies in huge volumes of data to train on: In our case, the algorithm would analyze news headlines and social media captions to try and see the correlations between texts and the meanings behind them. Given enough training material, the algorithm can“learn”(hence the name, machinelearning) about the context around the given text.
David Wallach, the creator of various financial data scrapers, echoes the shortcomings of traditional (non-deep learning) algorithms:
One main objective of this project is to classify the sentiment of companies based on verified user’s tweets as well as articles published by reputable sources. Using current (free) text based sentiment analysis packages such as nltk, textblob, and others, I was unable to achieve decent sentiment analysis with regards to investing.
For example, a tweet would say Amazon is a buy, you must invest now and these libraries would classify it as negative or neutral sentiment. This is due to the training sets these classifiers were built on. For this reason, I decided to write a script (scripts/classify.py) that takes in the json representation of the database downloaded from the Firebase console (using export to JSON option) and lets you manually classify each sentence.
We now see the importance of data in the sentiment analysis workflow. But how can we acquire it?
Overview of web scraping
In the term“sentiment analysis”,the “analysis”part refers to understanding the data — and the NLP algorithms we’ve explored earlier in the article can do just that.Web scraping, on the other hand, allows us to actually obtain the data to analyze.
Vladimir Fomenko, founder & CEO of Infatica.io
This term refers to the process of extracting and organizing data from websites.
How does web scraping work?
Web scraping is possible thanks to the way that websites organize data. Each website element — text, link, image, dynamic functionality, and so on — belongs to its respective category, denoted by standardized HTML tags.
A web scraper can navigate these elements with ease, locating and saving the data you need to gather.
NLP applications in FinTech
For example,Stocker, software for scraping financial data, follows the processes we outlined above:
It generates google queries, grabbing the latest articles that focus on a particular company.
Then, it parses the articles for information, trying to detect whether important pieces of information are positive or negative.
We can also use sentiment analysis in other areas:
Credit score analysis. Software product called LenddoScore can process the data available about the applicant online: This may include their social media profiles, browsing behavior, browsing history, and other markers. The software then rates the borrower’s creditworthiness.
Contracts analysis. JP Morgan hasimplementeda plethora of machine learning algorithms for numerous tasks. The company tested an NLP algorithm designed for contract analysis — and it has managed to save 360,000 man-hours in a year.
Customer service. Chatbots, the trendiest technology of the last few years, are powered by NLP algorithms. Financial institutions often pride themselves in offering great customer experience — and scaling their support via chatbots is a great way to do it.
Using proxies to ensure that your analysis runs successfully
Most websites don’t allow web scraping for various reasons. Here’s a typical example: a price aggregator tries to collect price data from multiple e-commerce businesses. Once this data is published on the aggregator website, potential customers will see thatVendor Moffers the best price. To prevent this, other vendors may restrict scraping their websites whatsoever.
Upon receiving a request to their website, they try to detect whether it comes from a genuine user or from a web scraping bot. While the genuine user gets a pass, the bot gets blocked.
However, itispossible to circumvent these anti-bot systems: using proxies, you can make your scrapers appear as real users.
Out of all the numerous proxy types,residential proxiesare the optimal solution: as their name suggests, they allow your scraper to appear as a real user, a resident of the country you selected. This enables you to bypass anti-scraping systems.
Every trader decides which type of analysis to use and which trading techniques to implement. But to my mind, improving financial sentiment analysis with AI and proxy servers is the new word in trading.
Vladimir Fomenko is the founder & CEO ofInfatica.io, a global peer-to-business proxy network
If someone reports sick after a gathering, a new AI-based system can trace contacts across four days and thousands of people in just four seconds.
This week, UNLV Lee Business School awarded Volan Technology the coveted Lee Prize Nevada Innovation Award for its advanced solution for enterprise-scale, precise and private contact tracing software. The technology could enable hospitality operators to make dramatic improvements in virus prevention—and save millions of dollars in manual tracing.
Halloween is just around the corner, waiting to surprise us. Though this year’s celebrations might be less extravagant, requiring special care and precautions, nonetheless we all know that Halloween traditionally has welcomed a slew of joyous activities including pumpkin carving, wearing scary costumes, elaborately decorating our homes, and so on.
The highlight, undoubtedly, has got to be the annual thrill of going trick-or-treating.
Perhaps you remember as a child going door-to-door in your neighborhood and the excitement at approaching the porch of a house covered with cobwebs and ghosts. Do you dare make your way to the front door? What goblins and other frights might await you? And, upon bravely knocking on the imposing door, recall the absolute delight at being given a chocolate bar or your favorite bubble gum.Off you would sprint, heading to the next house on the block.
Now, as an adult, hopefully, you either are the one dispensing those candies when impressionable youngsters knock at your door, or maybe you will be going outdoors and walking along with your children as they experience the same thrills that you did in your youth. It is fun to relish memories and also look toward future Halloweens too.
Speaking of the future (notice that seamless segue), some readers have asked me to comment on Halloween and self-driving cars.
I realize that your first thought might be that there is no particular connection between Halloween, one of the most revered celebrations each year, and the advent of AI-based true self-driving cars, an amazing technological innovation that is gradually emerging. It might seem odd to consider that there would be any type of connection between these two seemingly disparate facets.
Surprise!There is indisputably a means to connect the two, very much so.
Note: I hope that my written yelling of “surprise” at you did not startle you, though if you are reading this on Halloween, consider it the equivalent of being frighteningly startled with a spooky but festive boo.
The question for today’s discussion is this: What kinds of impacts might the prevalence of AI-based true self-driving cars have upon Halloween and our festivities thereof?Great question and I’ll get to it, but first, let’s make sure we all concur on what is meant by referring to AI-based true self-driving cars.
As a clarification, true self-driving cars are ones that the AI drives the car entirely on its own and there isn’t any human assistance during the driving task.
These driverless vehicles are considered a Level 4 and Level 5, while a car that requires a human driver to co-share the driving effort is usually considered at a Level 2 or Level 3. The cars that co-share the driving task are described as being semi-autonomous, and typically contain a variety of automated add-on’s that are referred to as ADAS (Advanced Driver-Assistance Systems).
There is not yet a true self-driving car at Level 5. We don’t yet even know if this will be possible to achieve, and nor how long it will take to get there.Meanwhile, the Level 4 efforts are gradually trying to get some traction by undergoing very narrow and selective public roadway trials, though there is controversy over whether this testing should be allowed. We are all life-or-death guinea pigs in an experiment taking place on our highways and byways, some point out.
Since semi-autonomous cars require a human driver, the adoption of those types of cars won’t be markedly different from driving conventional vehicles. But it is important that the public needs to be forewarned about a disturbing aspect that’s been arising lately, namely that despite those human drivers that keep posting videos of themselves falling asleep at the wheel of a Level 2 or Level 3 car, we all need to avoid being misled into believing that the driver can take away their attention from the driving task while driving a semi-autonomous car.
You are the responsible party for the driving actions of the vehicle, regardless of how much automation might be tossed into a Level 2 or Level 3.
For Level 4 and Level 5 true self-driving vehicles, there won’t be a human driver involved in the driving task.All occupants will be passengers.The AI is doing the driving.
Let’s begin with the important and assuredly upbeat insight that by removing the need for a human driver, there will no longer be any drunk drivers or DUI drivers on our roadways (well, at least with respect to the self-driving cars, though keep in mind that conventional human-driven cars will likely also still be on the highways and byways too).
What does it mean that there won’t potentially be intoxicated drivers during Halloween (or, at least a lot fewer ones)?In short, you can generally cross off your list any of those ghastly car crashes that lead to injuries and fatalities due to those out-of-their-mind human drivers.
This is especially noteworthy on the evening of Halloween since there are zillions of young children at risk that night. Kids love to dart out into the street on Halloween while trick-or-treating. They aren’t thinking about cars, they are thinking about the next house that has that inflated menacing ogre and beckons to them via the screeching sounds emanating from the blaring speakers placed in the bushes of the front yard.
Children on Halloween tend to falsely believe that the world has granted them a free pass to run and scamper all around the community.Though that would be a nice ideal, the reality is that there are today those drivers that insist on driving on the evening of Halloween and getting mired into the realm of where kids think they can go. One has to sympathize with the drivers that are sober and have little choice to drive that evening, perhaps to take their children to a Halloween party or for other needed purposes. Unfortunately, the bad apple drivers can spoil the whole barrel, as it were.
Plus, even a cautious and fully aware driver is bound to find themselves unnerved that evening. You need to drive slowly, really slowly, and this is a hard thing for many drivers to do. They are accustomed to driving at normal speeds and when driving a reduced speed it seems as though their vehicular movement is glacial in nature. Add to this the potential pressure of needing to get to someone’s house by a certain time, and you have an adult that can readily misjudge the roadways, leading to a calamity.
Bottom line: The more self-driving cars on the roads, the reduced chances of a human driver in that car that otherwise might have messed-up, one way or another, for whatever reason.
We can stick for the moment with the advantages and benefits side of the self-driving car versus human-driven car equation, and then, once I have covered most of those key points, we’ll consider some downsides too.
Pretend that you are busy at your home while handing out candy and also hosting your own adult-oriented Halloween party. Turns out that your teenage son or daughter was invited to a Halloween activity at the school grounds, but that is several miles away. Teachers at the school will be watching over the teens and you feel comfortable that your offspring will be safe there on their own.
All you need to do is drive your eager youngster to the school.
Instead of you being the driver, you could use a self-driving car to get your child over there. You would of course first make sure your child gets safely into the self-driving car, perhaps one that you own or that was available via a ride-sharing network, and the AI driving system then proceeds to drive over to the school. Once the event is completed, the self-driving car gives your offspring a ride back to your home.
This frees you from having to make the drive. Also, if you didn’t already own a car, or if you didn’t have a driver’s license, the use of the self-driving car solves several issues when desiring to provide your child with a lift to the Halloween event (and, once again, aids in preventing a potentially tipsy driver from getting behind the wheel).
Admittedly, some people repulsively recoil at the notion of having children riding in a self-driving car without any adult supervision. This will never-ever happen some parents exhort fervently. The idea is rather foreign to us currently and seems unimaginable, but we should be cautious in extending today’s cultural norms for what we might accept in the future.
Let’s continue our tour of the ways that self-driving cars will impact Halloween.
Some cities and suburbs have increasingly been setting up areas that allow for a drive-thru Halloween activity (especially due to the pandemic). You and the family pile into your car, and drive over to a park or parking lot that has a kind of haunted mansion or haunted city, as it were, created for providing a fun and spooky experience. Realize that you do not get out of your car. Instead, you remain in the vehicle, as though driving through a fast-food eatery, but in this case, it is an outdoor area set up with Halloween scenery.
If you were driving the car, it would likely be hard to fully relish the festive experience since you would be constantly having to watch where you are driving. Via a self-driving car, you would let the AI do the driving. This means that you and the rest of your family can all enjoy together the Halloween festivities, and nobody needs to be worrying about the driving.
Another somewhat new approach to Halloween that has been getting recent attention consists of trunk-or-treating.
Never heard of it?
You put Halloween decorations on the trunk of your car. Inside the trunk, you put bags or buckets of candies. When ready, you drive around the community, coming to a stop here and there, allowing kids to obtain their Halloween treats directly from the trunk of the vehicle. As to whether you get out of the car to dispense the candies, this depends (some purposely do not get out of the car as an added pandemic precaution for themselves and the kids that come to the car to retrieve the candies).
Not everyone likes this trunk-or-treating phenomenon.
Some point out that with Halloween candies dispensed from a house, you know where to go if there is something untoward handed to a child. The house is permanently affixed. On the other hand, someone driving a car around could be just about anybody, and they might not readily be traceable (for those of you that want to argue this point, it is true that you could copy down the license plate and trace the vehicle, but that’s a far cry from the aspect that a house is pinned to one readily known spot).
Anyway, whether you like or hate the idea, it perhaps is apparent that a self-driving car could enable such an approach if desired.
Yet another possibility for Halloween celebrations is the veritable Halloween car parade.People deck out their cars with Halloween banners and decorations. They put on costumes too if wishing to do so. You and your friends or family then get into the car and drive with other cars in a type of makeshift parade. This conga line of Halloween celebratory cars makes its way throughout the neighborhood. Horns are honked, people inside the cars are making noises befitting Halloween, and kids line the sidewalks, watching as the parade goes past their homes. As you might imagine, this is being spurred partially due to the pandemic, allowing people that already live together to be grouped into their car, and yet going outdoors to celebrate the evening.
One concern that some have about these Halloween parades is the possibility that some drivers will be drinking or have already had a few too many before deciding to join the car procession. Without seeming like a broken record that repeats itself unduly, if those cars were self-driving cars then the parade could meander unabated and without fear of a human driver doing something that could be injuriously unseemly.
Having covered the essence of the presumed upbeat or positive aspects of Halloween and self-driving cars, we now turn our gaze toward the less-so elements.
A perhaps obvious aspect about the advent of self-driving cars on Halloween is that it would allow adults to go to bars or parties and get smashed, if they wanted to do so, and not be held back by having to be a designated driver. Actually, this is true for any evening on any day of the week. You cannot fault the self-driving cars for this human behavior, but nonetheless could be a reaction by humans to the ease of having self-driving cars available.
Will self-driving cars spur people to drink or get drunk?
Nobody knows, and we won’t likely know until the day arrives of a prevalence of self-driving cars on our roadways.
Another aspect is the difficulty of driving on the roadways during Halloween.
Yes, even self-driving cars are going to find this to be a challenging driving task.
Do not falsely assume that merely because the self-driving car is using AI and has a collection of state-of-the-art sensors such as video cameras, LIDAR, radar, ultrasonic, etc., that it will perfectly and unerringly ensure that nobody is ever hit or hurt by a collision.
I’ve stated categorically and repeatedly that the notion of zero fatalities for self-driving cars has a zero chance of occurring. Physics belies such a belief. If a child darts unexpectedly from between two parked cars, and a self-driving car (or even a human-driven car) is cruising down the street, there might be insufficient time to stop the car before striking the suddenly appearing child. That’s a fact of physics.
When I mention this point, those in the self-driving industry are apt to instantly object. Therefore, let me be clear, I am not suggesting that self-driving cars will be less safe than human drivers. In fact, the expectation is going to be that self-driving cars have to be safer than human drivers. Thus, in the aggregate, we are presumably going to have many fewer injuries and fewer fatalities once we have a preponderance of self-driving cars. My point is that realistically it won’t go to zero. There will still be some non-zero number, though hopefully less than, a lot less, in comparison to the 40,000 annual car crash deaths in the U.S. annually and the approximately 2.5 million injuries.
Anyway, back to the point that on Halloween, especially so, the number of children and adults, perhaps even scampering dogs and cats, upon the roadways can be much higher than what normally is seen on the streets. This means many more objects that the AI needs to detect and discern as to which way the “object” is going and what it will do.
Children are small in stature and thus tend to be harder to detect. They might be wearing costumes that make their shape irregular in comparison to the expectations of what a person usually looks like. The kids can be erratic in where they go and whether they are sprinting or walking, or perhaps even crawling on the ground. All the kids and adults might be quickly stepping off a curb or willing to run amongst the cars that are making their way down the street.
Amidst all of this, it is nighttime and dark out, so the lighting of the scene can be quite problematic. Indeed, children might be carrying flashlights or lasers that they point at the cars, of which the sensors could be hampered by such actions.
In the self-driving car field, there is a well-known dictum that entails dealing with edge problems, also known as corner cases. Essentially, those developing self-driving cars are prioritizing what needs to be accomplished, of which just safely having the AI drive from a house to a grocery store during daylight is a keystone task. Unusual driving scenarios are labeled as being an edge or corner case, meaning that they are oddball or unique situations and presumably can be dealt with at a later time. The rule-of-thumb is to get the core stuff done first, and then worry about the rest later on.
It is safe to say that Halloween is an edge or corner case.
How many times a year do we all wander out into the streets, at nighttime, in costumes, with children aplenty?
Unless you live in an especially party-vigorous locale, the answer would seem to be that it is a once a year occurrence.
Even though Halloween is reasonably classified as a once-a-year instance, i.e., an edge case, this does not obviate the need to ultimately cope with the zaniness of the driving situation that arises.
Imagine if self-driving cars were unable to sufficiently drive around on Halloween, such that all self-driving cars had to be self-grounded that night. Assuming that people had become dependent upon the use of self-driving cars, it would seem inappropriate to have none available on any particular night, such as Halloween.
There is no doubt that ultimately self-driving cars will be enhanced to handle the particulars of a Halloween driving scenario. Also, to clarify, it is not as though self-driving cars of today could not likely manage to drive around on Halloween evening, it is just that out of an abundance of caution, it would seem unwise to do so at this time.
Consider what might happen if some self-driving car did ram into a child on Halloween (for sake of discussion, assume luckily the child is completely unhurt per se), this would generate unbelievably big news and become the headlines seen or heard all around the world. It would be a public relations nightmare that would exceed any of the scariest things you might ever envision associated with Halloween.
Let’s aim to avoid that kind of ghastly affair.
In any case, when thinking about the future, someday, you’ll be able to park your broom and let the AI do all the driving for you.
Happy Halloween to all and please be safe!
Copyright 2020 Dr. Lance Eliot
This content is originally posted on AI Trends.
[Ed. Note: For reader’s interested in Dr. Eliot’s ongoing business analyses about the advent of self-driving cars, see his online Forbes column: https://forbes.com/sites/lanceeliot/]
Google has had an eventful couple of weeks, announcing enhancements to its search and map capabilities at its virtual “Search On” event on Oct. 15, and on Oct. 20 being accused by the US Justice Department of engaging in anti-competitive practices in order to preserve its search engine business.
At the Search On event, Google detailed how it has tapped AI and machine learning techniques to make improvements to Google Maps as well as Search.
In an expansion of its search “busyness metrics,” users will be able to see how busy locations are without identifying the specific beach, grocery store, pharmacy or other location. COVID-19 safety information will also be added to business profiles across Search and Maps, indicating whether the business is using safety precautions such as temperature checks or plexiglass shields, according to an account in VentureBeat.
An improvement to the algorithm beneath the “Did you mean?” features of search, will enable more accurate and precise spelling suggestions. The new underlying model contains 680 million parameters and is said to run in less than three milliseconds. “This single change makes a greater improvement to spelling than all of our improvements over the last five years,” stated Google head of search Prabhakar Raghavan in a blog post.
In an improvement called Passages, an individual passage can be served up in response to a question, instead of a single web page. It does this by assessing the relevancy of specific passages, not just the overall page, helping to find “needle-in-a-haystack” information. Raghavan suggested the improvement will improve 7% of search queries worldwide.
In an example, a person asked, “how can I determine if my house windows a UV glass?” A ‘before’ screen on a list returns some website links; the ‘after’ screen on the right offers specific instructions, to hold a match up to the window.
“We’ve applied neural nets to understand subtopics around an interest, which helps deliver a greater diversity of content when you search for something broad,” Raghavan stated.
Data Commons Project Data More Plugged into Search
Google has been working since 2018, on the Data Commons Project, an open knowledge database of statistical data started in collaboration with the US Census, Bureau of Labor Statistics, World Bank and other organizations. New improvements make the information more accessible through Google Search.
If the user asks, “how many people work in Chicago?” the modified version taps into the Data Commons dataset. “We use natural language processing to map your search to one specific set of the billions of data points in Data Commons to provide the right stat in a visual, easy to understand format,” Raghavan stated.
Google Maps, in use since 2016, gathers its “busyness” insights from analyzing, aggregating and anonymizing location history data from users who have opted to turn the setting on in Google Account. The data is used to calculate the busyness of a place for every hour of the week. “The busiest hour becomes our benchmark—and we then display busyness data for the rest of the week relative to that hour,” stated Matt D’Zmura, Software engineer of Google Maps, in an account in Analytics Insight.
The “busyness” information will surface in directions and on maps, so that users do not need to search for a specific place to see how busy it is. The features will soon be available to Android, iOS, and desktop users worldwide, Google indicated.
On the business side, the outlook for Google’s digital advertising business remains an issue amid the coronavirus pandemic, according to a recent account in Investor’s Business Daily. A rebound in Google’s core internet search advertising business is key for the stock, stated Bank of American analyst Justin Post in a recent report to clients. Google has a scheduled quarterly earnings report on Oct. 29.
YouTube’s ad revenue growth came in under expectations in the June quarter while cloud computing revenue growth slowed from the previous quarter, Post reported.Google has struggled to gain share in cloud-computing services against Amazon and Microsoft, however Google has strength in its AI-spanning digital advertising, Google Cloud Platform, YouTube and its search technology businesses, the analyst indicated.
Justice Department Alleges Anti-Competitive Practices
The bad news from the Justice Department came in the form of an antitrust lawsuit alleging Google uses anti-competitive tactics to preserve a monopoly for its flagship search engine and related advertising businesses. It was called the most aggressive US legal challenge to a company’s dominance in the tech sector in more than two decades in an account in The Wall Street Journal.
The government alleges that Google’s practice of paying smartphone manufacturers and browser suppliers including Apple (with Safari) to maintain Google as their preset, default search engine, creates an unlawful, self-reinforcing cycle of dominance.
“Google achieved some success in its early years, and no one begrudges that,” stated Deputy U.S. Attorney General Jeffrey Rosen. “If the government does not enforce its antitrust laws to enable competition, we could lose the next wave of innovation. If that happens, Americans may never get to see the next Google.”
Google’s lawyers countered that they see the lawsuit as deeply flawed. “People use Google because they choose to—not because they’re forced to or because they can’t find alternatives,” stated Kent Walker, Google’s chief legal officer. “Like countless other businesses, we pay to promote our services, just like a cereal brand might pay a supermarket to stock its products at the end of a row or on a shelf at eye level.”
The last similar US government antitrust case against a major US technology firm was US against Microsoft in 1998, the same year Google was founded. The Microsoft suit eventually ended in a settlement. The Google case is likely to take years to resolve.Facebook, Apple, and Amazon are also facing antitrust scrutiny from Washington.
Google owns or controls search-distribution channels accounting for about 80% of search queries in the US, according to the lawsuit and third-party researchers. The government states that effectively leaves no room for competition, resulting in less choice and innovation for consumers, and less competitive prices for advertisers.
The Journal reported on tension within the Justice Department about the timing of the suit against Google, with some staffers suggesting Attorney General William Barr was driven by an interest in filing before the Nov. 3 presidential election.