The Algorithm Empire, some examples of algorithms that manipulate users (Part # 2).
Originally Posted: Article.Coinpayu
In the first part of this article I commented and I quote: "The algorithm is the king of the Internet, the one that controls what we see, what we read, what we listen to and what we share. The one that knows us better than ourselves and the one that It influences us without us realizing it. The one that creates a custom information bubble for us, where only what we like, what confirms us and what suits us reaches us.", "The algorithm is the empire that dominates us, the invisible power that conditions us, the silent threat that stalks us. Can we escape from it? Can we resist it? Can we control it? Or rather, can we control ourselves?" But to "defend" ourselves from them you have to start by knowing them, some examples:
π Recommendation algorithms: these are programs that use artificial intelligence to suggest content, products, or services to users based on their preferences, interests, or behavior. These algorithms are based on the analysis of the data that users generate when interacting with a platform, such as their searches, their purchases, their ratings or their comments.
π Search algorithms: they are the ones that order and display the results of the queries we make on Internet search engines, such as Google or Bing. These algorithms can influence what we see, read, and learn, as well as what we buy, vote for, or believe. These algorithms may be biased by commercial, political or ideological criteria, and may hide or privilege certain sources of information.
π Advertising algorithms: they are the ones that select and display the ads that appear to us on web pages, social networks or mobile applications. These algorithms can influence our consumption decisions, our emotions and our privacy. These algorithms are based on our personal data, our browsing habits and our interests to show us personalized and persuasive ads.
π Moderation algorithms: they are the ones that filter and eliminate content that violates the rules or policies of digital platforms, such as social networks or messaging services. These algorithms can influence freedom of expression, public debate and the quality of information. These algorithms can make mistakes, censor legitimate content, or allow harmful content.
In order not to extend the article too much, I will break down only some functions of the recommendation algorithms (the most common):
π Content-based filtering: this method consists of recommending content similar to what the user has already viewed or purchased. For example, suggest movies or series similar to those recently watched on a streaming platform. This method is based on the characteristics of the content and the user's profile.
π Collaborative filtering: this method is based on the analysis of relationships between users to recommend content. Based on ratings, reviews or purchases, the algorithms can identify users with similar preferences and recommend content that other similar users have liked. This method is based on the similarity between users and the popularity of the content.
π Machine Learning: This technique predicts user preferences based on their past behavior. Algorithms continually learn from the data collected and can offer more relevant and personalized recommendations. This method is based on learning user patterns and adapting to their changes.
Another important aspect is the bias in these recommendation algorithms, which is nothing more than the tendency of these programs to favor or discriminate against certain content, products or services based on criteria that are not objective, fair or transparent. These criteria may be related to gender, race, age, ideology, religion or any other characteristic of users or content. Bias in recommendation algorithms can occur for different reasons, such as:
π The definition of the problem: recommendation algorithms are designed to meet a specific objective, such as maximizing sales, dwell time or user satisfaction. This objective can influence the type of content that is recommended and how its relevance or quality is evaluated.
π Data collection: recommendation algorithms are based on data collected about users and content. These data may be incomplete, out of date, unbalanced, or biased by external factors. For example, if more data is collected from one group of users than from another, the algorithm can learn to better recommend that group and ignore or disadvantage the other.
π Data analysis: recommendation algorithms use mathematical and statistical techniques to analyze data and find patterns or correlations. These techniques may have limitations, assumptions, or errors that affect the results. For example, if a similarity measure is used that does not capture well the diversity or complexity of users or content, the algorithm may recommend content that is too homogeneous or irrelevant.
Bias in recommendation algorithms can have negative consequences for both users and content providers. On the one hand, it can affect the quality, diversity and fairness of the recommendations that users receive. On the other hand, it can affect the visibility, reputation and profitability of the content offered by providers. Bias in recommendation algorithms is a difficult problem to detect and correct, since it depends on many factors and can vary depending on the context and the moment. Furthermore, recommendation algorithms are often opaque and complex, making them difficult to understand and explain. Therefore, a joint effort of researchers, developers, regulators, and users is required to identify, measure, and mitigate bias in recommendation algorithms.
Lastly, as I expressed in the previous installment: "The algorithm exists because we exist, it changes because we change, it dominates us because we let ourselves be dominated by it, but we can also dominate it. We just have to want it. AND DO IT."
Author's Note: The opinion expressed here is not investment advice, is provided for informational purposes only, and reflects the opinion of the author only. I do not promote, endorse or recommend any particular investment. Investments may not be right for everyone. Every investment in the market and every trade you make involves risk, so you should always do your own research before making any decision. I do not recommend investing money that you cannot afford to chair, as you could lose the entire amount invested.
* Tools, Platforms and Applications: Earn active and passive earnings (using referral programs). However, I recommend that you do your own research.
π Giveaway 100.000.000.000 $LOVELY - Participate and share your referral link. Luck may be on your side. End: 20.8.23.
π² AddmeFast - Tools to promote and increase the sources of traffic, visibility, reach and reputation of your social networks. Also earn money and/or cryptocurrencies.
π² CoinInn - Trading - Bonus (π) between $500 and $750, only profits are withdrawable (Need to complete KYC to extract).
π² StormGain - Trading - They can start without investment, capital is acquired with the Bitcoin Cloud Miner. Only profits are withdrawable.
π² QuantFury (JRRU2593) - Trading - Bonus (π) in cryptocurrencies or company stock worth up to $250.
π² CoinEX, BingX, Doex, PrimeXBT, KuCoin, Cex, MoonXBT, BiBox, BitYard, ZooMex - Trading - They offer different bonuses.
π² CoinMarketCap, TradingView - Indispensable tool for traders.
π² Publish0x - Earn daily ETH, AMPL & $POT for reading or writing articles. Similar to Bulb & Article.CoinPayu.
π² Bulb - Earn daily Crypto (BULBTokens) and NFTs (BULBmoji) for reading or writing articles and interacting with publications (Airdrops). Similar to Publish0x & Article.CoinPayu.
π² Article.CoinPayu - Similar to Bulb & Publish0x, Earn BTC.
π² Coinpayu, Cointiply, Freebitco, AdBTC - Generate additional income without investment.
π² Bitrefill - Living with crypto, a philosophy of financial freedom. Travel, play, eat and live with BTC.
π² Nowpayments - Non-custodial cryptographic payment gateway that allows merchants to accept payments in more than 50 cryptocurrencies.
π² Ledger, BixBox02, Ellipal Titan, Keystone - Hardware Wallet with affiliate marketing.
π Special +18 (Cell Phones Only) - Dating, Sweepstakes, Health, Games, Horoscope, Video, Download, Others, Special +18 Ramdom.
π Mainstream (Cell Phones Only) - Dating, Sweepstakes, Health, Education, Games, News, Music, Video, Horoscope, Download, Recipes, Others, Mainstream Ramdom.
π―βπ― - Products & Others.
"You will ask yourself: And if I take a risk and lose...? I will ask you: AND IF YOU RISK AND WIN? Success begins with thought, because sooner or later the man who wins is the one who believes he can do it. Do not be afraid of mistakes or failure, winners are not afraid of losing, losers are, in most cases the risk comes from not knowing what you are doing, so trust yourself, learn, be patient, manage your emotions and above all, enjoy the journey, what the wise man does at the beginning, the fool does at the end" - Anonymous.
πβ Buy Me a Coffee βπ
