by Sentimenti Team | Feb 3, 2023 | Market research, Politics and Social
Disinformation – a word that appears in the public space every day. Russia almost every day conducts an intensive disinformation campaign against not only Ukraine, but also Poland and countries that help Ukraine, both militarily and humanitarian. Every day we face a séance of hostility, but above all a flood of lies about us.
Disinformation and pro-Russian narrative on X (formerly known as Twitter). What is being written about?
Analysts and specialists in disinformation and cybersecurity agree here. The purpose of the online campaign, which has been ongoing since the beginning of Russia’s invasion of Ukraine, is primarily to provoke hostility and antagonize Ukrainians and Poles. With each month of the war in Ukraine, the scale of the attacks intensifies, and increasingly harsh accusations and threats are made against Poland.
According to Stanislaw Zharyn, who has recently been acting as the government’s plenipotentiary for the security of Poland’s information space, Russia is actively conducting a psychological operation against Poland and the West as a whole. The operation is aimed at intimidating and forcing submission to Russian imperialism.
Russian propaganda is constantly conducting online operations aimed also at intimidating Poles. On the Internet, all themes appearing in Russia’s popular propaganda programs are being expanded. We decided to check what kind of content is created by the most popular Twitter accounts known for sowing disinformation, fake news about the war and – generally speaking – for their love of Russia.
We selected the 5,000 most popular posts (with the highest number of accumulated interactions) posted to this channel from each of the months between January 2022 and January 2023, i.e. for a full 13 months. A total of 494 profiles were manually monitored. The data collected in this way was further examined for emotion and sentiment, and we further examined which threads and thematic contexts aroused extreme emotion and generated high emotional arousal.
Among the online personalities we surveyed are mostly anonymous accounts and profiles, which often make typical Russian mistakes. They are created in part by transferring Russian content to Poland using automated free translation programs. Unfortunately, more influential people are increasingly included, spreading pro-Kremlin propaganda as a whole.
Festival of fake news and disinformation brings results
The incentive and starting point for our analysis was an article that appeared on gazetaprawna.pl (Polish). As we read in the material, titled “The Russian propaganda. “Poles increasingly susceptible to Russian propaganda [SURVEY],” recently there has been a strong increase in the number of people who more or less consciously share the theses circulated on the web that are part of pro-Russian disinformation. The Warsaw Enterprise Institute conducted its survey in this regard.
The results of the aforementioned survey show the same thing we see in the data we collected. Contexts of Ukrainian refugees, who are referred to as economic migrants, appear regularly. Even more frequent references are made to history, namely Volhynia and Bandera. Propaganda penetrates from Russia to Poland without any barriers via social media, but not only.
Returning to the 494 X (formerly known as Twitter) profiles we took under our magnifying glass, one thing strikes the eye above all. This is the extremely high level of activity. Some of the Twitter accounts are active to such an extent that it raises serious doubts about whether there are actually people behind them.
The Polish Anti-War Movement and the disturbing results of public opinion polls
The content is mass-collected in such quantities (200 or more posts a day) that it is challenging to imagine that the people behind it have anything else going on in their lives. There are also accounts that can flood a social network with several thousand pieces of content in a month. As for month-on-month increases in activity, for some periods the activity even increased by more than 80%. The trend is also noticeable and unambiguous.
What can be seen in the chart is, of course, only a slice of what is actually happening online. The point, however, is that this type of activity brings tangible results to those who conduct such a campaign online (increasingly, they also conduct it offline, i.e., on the streets of Polish cities).
The conclusions of the WEI study, as reported in Gazeta Prawna, are clear. In September 2022 and January 2023, the survey was conducted on 1,061 Poles. A comparison of the collected information yielded grim conclusions. , In January this year, as many as 41 percent of respondents (previously it was 34 percent) “strongly agreed” or “tended to agree” with a minimum of four statements characteristic of the Russian narrative
The opinions of Poles in this survey are very consistent with what narrative is served to them online, among other things. For 63 percent, Poland cannot afford to take in refugees. 41 percent are also convinced that refugees from Ukraine are economic migrants. 30 percent of Poles believe that we should not help Ukraine unless they apologize for the difficult history of relations between Ukraine and Poland. The group of Poles who believe that Russia should not be teased because it is a nuclear superpower has increased to 40 percent.
Content colluded by pro-Russian profiles online. What kind of sentiment?
Pro-Kremlin propaganda is beginning to increasingly shape what is going on in the minds of Poles. Declining support for supporting Ukraine or welcoming Ukrainians fleeing the war is a phenomenon that is occurring in parallel with what is happening to our economy.
Poles are increasingly connecting the dots and believe that the war is to blame for the fact that we have less and less in our wallets. This thesis is also being used in the narratives being spread.
The chart below is an analysis of the content we’ve gathered from nearly 500 profiles that colportage constantly pro-Kremlin content on X (formerly known as Twitter). Sentiment averaged negative on a monthly basis throughout the period analyzed, which is not surprising
However, not much variation in sentiment was noted for the entire 13-month period under study. What is clear, however, is how the narrative changed after the start of the war in February 2022, and that there was a marked jump in negative intensity.
Contexts and topics covered by disinformation accounts
When it comes to the content of posts uploaded to the web, we checked the ones that carry the greatest emotional charge. It’s all about anger or revulsion. The following analysis is the result of this. The different colors denote groups of topics. The color black identifies political topics. The red color is in contexts related to the war in Ukraine and derivative topics such as the Volyn slaughter, genocide and the military.
The yellow color refers to issues related to refugees from Ukraine. Blue tinges the topics related to energy and climate policy and raw materials. We were also able to isolate issues related to coronavirus, which we marked in green.
The analysis shows perfectly what the analyzed profiles are colporting. We can see what the intensity is in terms of topics and what issues are being covered. These accounts produce intensely content hitting the Polish government. They are also spreading disinformation in the field of health care or pandemics. We also see the topic of Ukrainian refugees and the impact of their presence in Poland on our economic situation.
Russian trolls will be colporteurizing more and more such information in the near future, as indicated, for example, by the first chart in this article. Soon we should expect – in connection with the planned and long-announced spring offensive – more uploads, but with strictly military topics (including mobilization in Poland, which is not present).
One can expect a difficult time ahead. It will be difficult to distinguish truth from lies, but the goal is invariably the same – to divide Poles and Ukrainians, Putin’s strategic goal.
by Damian Grimling | Oct 7, 2022 | Sentistocks
Credit Suisse’s stock price has been on the decline for many months. Confusion over the bank’s financial condition is growing. There are further reports in the media about the potential insolvency of the institution, which is the second largest Swiss bank. Could its share price have been predicted? Yes – thanks to artificial intelligence.
Safe as a Swiss bank. Credit Suisse CDS price is a real trouble
Credit Suisse’s stock price has been on the decline for many months. As Sentimenti deals with, among other things, predicting the prices of financial instruments or cryptocurrencies, we decided to take a look at the atmosphere and mood on the web around Credit Suisse.
The analysis looked at the level of intensity of emotion and emotional arousal among authors of online opinions and comments, which were written in English and German. That’s a total of more than 40,000 mentions from May to September 2022, which our artificial intelligence looked at. The results are surprising.
It was not surprising to find a high level of intensity of emotions generally understood as negative emotions somewhat related to uncertainty (fear and sadness) or neutral emotions (surprise and anticipation). However, it was a big surprise to find an increasing level of intensity of emotions conceived as positive (trust and joy).
A detailed analysis of the content of the entries showed that the entries characterized by a high intensity of these favorable emotions are those expressing – generally speaking – satisfaction with Credit Suisse’s troubles. This state of affairs significantly affected the level of the correlation index of the intensity of emotions and the bank’s stock price.
The study of emotion diads (which form according to Pluchik’s theory) also leads to interesting observations, including showing how market attitudes changed. Thus, emotions correlated most strongly with Credit Suisse stock prices:
in May of expectation and trust – forming a diad of fatalism;
in June and July of fear and sadness – forming a diad of despair;
in August expectation and sadness – forming a diad of pessimism;
in September of expectation and disgust – forming a diad of cynicism.
Using Pearson correlation coefficient analysis, the interaction between stock prices and the intensity of emotion around Credit Suisse on the Internet was examined.
Credit Suisse problems. Are we in for a repeat of Lehman Brothers?
The analysis showed that there was a rare case where all emotions (including those considered positive) correlated negatively with Credit Suisse share prices. This meant that, in this case, a recorded increase in the intensity of emotions was linked to a decrease in share prices. In September, the strongest correlations with the share prices of Switzerland’s second largest bank were revulsion and expectation forming – according to Pluchik’s theory – a diad of cynicism.
A prototype stock price prediction model was fed for Credit Suisse with financial data (share prices) and data on the intensity of emotions, sentiment and emotional arousal around the company. A model built on an artificial neural network using BiLSTM (bidirectional long short-term memory) was used.
Using our predictive model (Sentistocks service), a success rate of up to 88% in predicting the trend of Credit Suisse’s share price over the period under study was achieved. This shows that potential investors could have successfully predicted the quotation solely on the basis of online discussions about the bank’s financial situation.
Investors remember the great collapse of Lehman Brothers. 14 years ago, it contributed to the outbreak of the global financial crisis. Will Credit Suisse face the same finale? We do not know. But the cost of CDS (insurance against bankruptcy) for the institution has approached the highest levels in almost 15 years. The bank itself foresees bumpy times ahead.
Credit Suisse CDS price. How emotions suggest the direction of a share price trend?
Sentistocks has developed effective predictive models for instruments in the cryptocurrency market. These models use both financial and emotive data to predict future prices (rates). The high success rate of the predictions developed with our tool confirms the huge role played by emotions in financial markets.
by Sentimenti Team | Mar 18, 2022 | SentiBrand
Customer service is not only about everyday relationships and contacts with the recipient of our products and services. It is also about constantly analyzing their needs, desires and finding ways to satisfy them. Due to the emergence of the COVID-19 pandemic, a large number of companies had to work remotely, which made identifying customer needs quite a challenge.
Customer service. Promotional tools that must be used
The COVID-19 epidemic has caused the term customer experience (CX) to gain importance. Analyzing and solving customer problems has become much more difficult. So how can we improve customer service in the new era, since we already know that there is no going back to what it was like before the epidemic?
There are many tools on the market that can be useful both in remote/online and direct contacts. Below we have selected the key and most helpful ones, the operation and method of use of which may be essential to improve customer service.
Media monitoring. Social listening tools and brand image
Social media monitoring tools are designed to listen to what is being said in social media about a company, brand or any other topic. The market offers many solutions in this area that are more or less tailored to the needs of companies, but they all undoubtedly have one basic feature in common. They allow you to search social media for mentions of a company/brand/product and extract not only what is being said, but also valuable information.
Internet and social media monitoring is perfect for analyzing the effectiveness of current campaigns, analyzing opinions and comments on new products and assessing the general reception of any decisions made by the company. But that’s not all. This type of software can also be used to check what our competition is doing. This will make it easier to compare their strategy with our own. This in turn results in the introduction of necessary changes.
Internet monitoring will also play a key role in avoiding or suppressing image crises that can occur anytime and anywhere. We are talking about media that live 24 hours a day, 7 days a week. Fortunately, social listening works in real time, so if any negative reactions appear online, we will be able to quickly locate the source of their origin and take concrete steps to nip the crisis in the bud.
Internet monitoring. Sentiment analysis and analysis of emotions in Internet user opinions. Insight marketing
Explaining this is a bit complicated, but sentiment analysis is essentially a tool for reviewing communication – whether it is emails, tweets or other – and extracting the information contained within it. However, it is more than just reading text. Sentiment analysis is the analysis of tones and contexts to best assess the meaning of words, rather than taking them at face value.
Sentimenti has an excellent tool to help analyze feedback. It also goes beyond the simple sentiment that companies have known so far, as it provides insight not only into whether a given opinion is positive or negative, but also to what extent. Additionally, the emotions contained in the customer’s comments are examined. There are as many as 8 of them.
Interestingly, Sentimenti tools, thanks to the use of artificial intelligence, examine tens of thousands of opinions in the blink of an eye as if over 20,000 people were reading and evaluating them. If one person did it, the analysis would take several or a dozen days. This is a revolutionary solution that has not been available on the Polish market so far. It is also based on science and is the result of months of expensive research on real people, not an artificial algorithm.
Sentimental analysis is particularly useful in markets where opinions change quickly and services must be constantly updated. However, any organization that wants to save time and respond to customer opinions faster and more effectively will also benefit from this analysis. It is good that ready-made software already exists for this type of analysis, which – considering the complexity – is much more convenient.
SMMW, behavioral analytics and consumer behavior research
Behavioral analytics involves studying consumer behavior on the web. The primary use of such tools is to predict customer needs and desires and proactively solve problems that may arise even before they occur. By checking user demographics, their previous purchases, and similar information, you can predict much of what general problems may arise. Using personal data, you can better understand their unique behavior.
A/B testing, which can be obtained from Google Optimize or a similar tool, is particularly important for checking the effects of changes to the page, while session replay tools such as Hotjar provide more in-depth insight. Behavioral analytics can also be used in sales, where cross-selling and personalization can provide a big boost in revenue by ensuring purchases in places where customers might not see the right products.
It is also useful for creating personalized marketing campaigns and tracking return on investment. Hotjar and Lucky Orange offer tools such as heatmaps, which are particularly helpful in this case. They will help you understand what your visitors are doing on your website, what they are clicking on, and what they may not be seeing.
Customer relationship management (CRM) systems. Customer segments. Ways of communicating on the internet
A customer relationship management (CRM) system is a relatively new technology on the market. The concept of CRM did not emerge until the 1980s. The idea behind CRM is to enable organizations to manage customer interactions and streamline processes to improve overall customer service. The goal is to increase profitability for the business, but it’s also a give-and-take approach that delivers real benefits not just to you, but to the customer as well.
The benefits to the business are wide-ranging. Sales teams can use CRM information to see how individual teams, campaigns, and products are performing. Marketing teams can use CRM data to predict the customer journey a particular person or demographic group might take and tailor their approach to the customer accordingly. Meanwhile, customer service agents can track cases efficiently across multiple platforms without the risk of misinterpreting queries.
Tools in action. Cloud software
Cloud software allows for a personalized approach to customers and extends its reach beyond anything that could have been expected just a few years ago.
Cloud computing not only allows organizations to track the effects of strategies and changes in real time, but also enables every department in the company to access information from anywhere. Key examples include Salesforce, a SaaS platform with useful marketing features, and file-sharing services like Dropbox and Google Drive.
This approach is similar to what is currently being used in scientific research. And for good reason. Cloud software allows teams across the company to instantly access any information in a database and use it to improve the customer experience. Instead of individual teams figuring out what works and what doesn’t, they can skip any mistakes that have already been tried and immediately improve the customer experience.
While this approach is more beneficial for large companies — also for cost reasons — there are advantages for smaller businesses as well. With a centralized place where all information is stored — online, without the risk of losing files or accidental corruption — teams can save time preparing for previous events.
What is omnichannel? Learn about omnichannel sales
Omnichannel often goes hand in hand with cloud databases. It’s simply a must-have if you have multiple channels of communication with your customers. This approach takes all of your contact lines and combines them into a single database, which varies from company to company. In some smaller organizations, a simple spreadsheet will suffice. In larger companies, a cloud database is often essential for cross-departmental and remote communication.
So what are the benefits of omnichannel for consumers? When communicating with someone about a query, no one likes to repeat themselves. Having easily accessible information that someone can access, even if it’s not the person who wrote the data, not only saves employees time and effort, but also prevents the customer from getting frustrated due to a lack of communication.
The best response is a quick response. Another thing is to provide the customer with all the necessary information without bothering them again.
Companies like Disney, Walgreens, and Starbucks have successfully implemented the omnichannel approach. And since tools like Salesforce Social Hub, TalkDesk, and Genesys Multi-Channel are available on the market and can help your business, why not give it a try?
Promoting your business online. Your company website and its constant monitoring
Google Analytics is a must. By that, we mean tracking user behavior on your company website, how they navigate it, what sections they focus on with their mouse, and what links they click on. This is about tracking actions, which is completely harmless and non-invasive, since the customer visiting your website has necessarily given their consent to interact with you.
Tracking your website can reveal a lot about your site’s layout, allowing you to identify any awkward elements that may need to be changed, showing how users navigate irregularly, and where they may end up leaving without making a purchase.
The data collected from tracking your website can be easily translated into a mapping of the online customer journey, thus deepening your understanding of how your customer thinks. While it won’t directly benefit your customers, the information you collect can provide insight into how to best design your website to make it easier for your customers to access the key information they’re looking for.
What’s on the market? Adobe Analytics offers traffic and multi-channel data collection, while Clicky analyzes real-time data on page views and events. Which tool you choose really depends on your business’s needs. Each offers the same basic tracking features with different analysis options.
Chatbots in Marketing and Process Automation
We left the topic of automation for last. It’s exactly what it sounds like: creating a system that does the job for you as much as possible. There are many ways to successfully implement it in customer service. For example, fully automated call centers or chatbots, which have become increasingly popular in recent years.
Some business leaders still scratch their heads at the thought of bots talking to their customers. But they’re very advanced and can be a great way to narrow down online queries — much like the aforementioned automated telephone switchboards.
More advanced versions can also tap into previous interactions or FAQ databases to find answers without involving a human agent. In this case, the benefit to the customer is speed, as a computer program can respond to any interaction almost immediately.
One of the most important benefits of automated responses is 24/7 support, not just during business hours. While there’s certainly no human touch that comes with talking to a team member, those needing help will prefer a bot’s response to waiting until the next morning to get valuable information that could ultimately be crucial to making a sale or turning a customer over to a competitor.
To sum up…
Customer experience is all about information. The more of it we have, the harder it is to find anything meaningful in it. However, with the right tools, which we wrote about below, you can let your computers do most of the work and still get much more meaningful analyses.
Speed is key these days when it comes to customer satisfaction, especially the younger ones. That’s why it’s worth taking a closer look at our advice and checking which of them meet the needs of companies.
by Agnieszka Czoska | Feb 7, 2019 | SentiBrand
Not only our analyses show from time to time that the more emotions we show, the more we receive from our interlocutors. What’s more, I think everyone knows the basic principle of advertising and marketing: emotions pay off because they generate reactions. On the wave of these beliefs, we decided to check whether actually popular Facebook and Twitter posts are more emotional than those that evoke less liking or commentary.
We used Facebook Insights and Tweet Activity analytics statistics. We focused only on those that tell us about the actual interest in a given post: the number of reactions (likes and comments). We ignored the number of page views and similar “big numbers”. – Emotions do not necessarily affect the algorithms of social media, but they should be the reaction of those who observe a particular channel. Where did we get the data from? Not from the Sentimenti website, we have too few fans so far. But we happen to be familiar with a certain pop-culture portal – the statistics are for posts from Nie Tylko Gry.
Emotions on Facebook
We have collected data from January 2019. The portal has more than 1700 viewers on Facebook, at that time it published about 3 posts a day, including multimedia (e.g. containing film posters), links to its pages, several links to films (mainly trailers). We did not analyze these types of posts separately, we would need much more material to do so. We focused only on the text. The posts evoked 20 likes or 23 reactions on average (3 comments on average). We filtered out the 20 most popular posts and the 20 with the least reactions.
In both types of posts we found less than 20% of emotional words, 15% in the most popular ones and 18% in the second group. This difference is not statistically significant, so the “amount” of emotions alone is not responsible for the popularity of the post.
If we look at the distribution of individual emotions, we can see that these two groups of posts differ most in the presence of sadness and trust. If we analyze 8 emotions together, the difference is not statistically significant, but already the set of trust, sadness and anger gives a result that shows the difference between the groups (Chi=13,945; p=0,002). Interestingly, in the most popular posts there are more negative emotions and less trust. Does writing about pop culture feed on criticism? Or maybe doubts about the quality of the promoted works?
Emotion on Twitter
We also analyzed the twitts of the portal, this time in December 2018 and January 2019. The portal has less than 200 observers on Twitter, published, as on Fecebook, about 3 posts a day: photos and links (mainly to its website). On average, they generated 4 reactions to the post (including: likes, comments and sharing). We filtered 60 of the most and least popular posts.
The percentage of words carrying emotion was slightly smaller for twitts, about 13%, and did not differ between the two groups of entries. Interestingly, longer twitts seem to be more popular than shorter ones. However, none of these differences are statistically significant.
Joy and trust seem to differentiate twitty most strongly. In fact, if we consider these two emotions and fear, we get statistically significant differences between the groups (Chi=8,569; p=0,014), while for 8 emotions the result is not significant and we can only talk about the tendency. In popular twitts trust is more often expressed, and less often – fear and joy.
Emotions in social media
As you can see, Twitter and Facebook posts from the Not Only Games portal we are analyzing show the opposite relationship between emotions and popularity, although on both platforms trust seems to be an important emotion for the audience. It is more important than the ratio of words that carry emotion to neutral. In the case of social media, therefore, the portal cannot simply talk about the influence of emotions on the popularity of an entry, no differences remain relevant if we compare all popular and less popular posts. We have to treat both portals as separate collections of texts to say something about evoking readers’ reactions.
What does this difference mean? Perhaps the author-reader relationship on Twitter and Facebook is different, but the administrator may also be key here – and for “Nie Tylko Gry”, someone completely different deals with each of these channels of communication with the recipients.
In this analysis we have shown that emotions, especially trust, influence the popularity of a post on a social network. At the same time, there is no uniform, valid for all media – on Twitter, positive emotions “won”, and on Facebook, negative emotions. It is also not enough to simply show any feelings – the proportion of emotional to neutral words was not important.
Getting to know your audience is crucial for promotion in social media. In order to encourage interaction with our post, we should find out what our readers, potential clients and supporters actually react to. Do they prefer a photo or text? Probably the first. But do they prefer joy, anger or trust? That’s what we won’t find out from the Facebook algorithms yet. As we have shown, these preferences depend on the communication channel and without reliable data analysis we will not be able to evaluate them.
