What’s going on with Facebook after the outage and whistle-blowing?

I did a Twitter analysis based on recent 6000 tweets. The text-mining results are provided below-

It appears from the following word cloud that it is not only the outage impacting this brand but also the whistleblower is impacting this brand image of Facebook.

Outage and whistleblowing impact on Facebook

From my analysis, I found that people are expressing surprise-type feelings related to Facebook on their tweets.

We can see from the following word cloud, these surprise feelings are related to WhatsApp, Instagram, Facebook, outage, and whistleblower.

Surprise Feelings related to Facebook

What shapes of robots are preferred by consumers in hospitality services?

From one of our research, we found that nonphysical robots, like artificial intelligence in smartphones (Please see photo 6 below), are the most easily identified and preferred shape (mean: 3.65). We can also see from the following pictures and statistics that trash bins (photo 2) or tray robots (photo 4) (mean: 3.49; 3.47, respectively) are also preferred by consumers in the hospitality industry. An interesting point is that Human-looking robots
(photo 1) appeared only one from the lowest preferred (mean: 3.35). It may be because it scares consumers and it requires more research to be done in the future to understand this situation. The spider shape robot was the lowest preferred (mean: 2.31) robot (photo 5) and it lacked hospitality service application.

In conclusion, we can say that human shape robots may not be preferred always in the service sector. Consumers may like nonphysical shape robots because they may provide consumers a sense of control instead of getting scared looking at their scary shapes such as spider robots.

To know more about this research, please visit our following published research in a TOP journal –

de Kervenoael R.;Hasan R.;Schwob A.;Goh E. (2020) ‘Leveraging human-robot interaction in hospitality services: Incorporating the role of perceived value, empathy, and information sharing into visitors’ intentions to use social robots’. Tourism Management, 78 [DOI][Details]

Text mining the tweets to understand peoples’ feelings towards a brand (Example: Ben & Jerry’s)

Is it possible to Text mine the tweets to understand peoples’ feelings towards a brand?

Yes, it is possible to understand what type of feelings people are expressing regarding any brand from the tweets of Twitter. I am going to demonstrate the example of Ben & Jerry’s below.

We have collected 6000 tweets from Twitter that use the keyword Ben & Jerry’s. the following picture is the word cloud result ( larger font size represents words more frequently used in this word cloud), which shows us what people are talking about Ben & Jerry’s.

Word Cloud of Ben & Jerry’s

We are going to demonstrate in the following picture what type of feelings consumers are expressing on Twitter related to Ben & Jerry’s.

People’s Feelings related to Ben & Jerry’s

From the above results of text mining, we can understand that people are expressing more joy and surprise feelings related to Ben & Jerry’s brand and it represents a good reputation for the brand.

Text Mining can be used to monitor social media

Measuring public opinion was always about extrapolating from surveys and hoping the small sample you selected was representative of the general public. Today, individual public opinion is ripe for the taking on Facebook, YouTube, Instagram, Twitter, Reddit, WhatsApp, and whatever new product review or augmented-reality platform pops up next.

With the advancement of Artificial Intelligence, we can now analyze texts using different types of software.

For one of my research projects, We have analyzed 40,000 Amazon Reviews of Smart Wearable Devices like Fit Bit, Samsung Gear 2, and 3. The Following output provides us an idea, what people are talking about your brands or products. It may be difficult to go through 40,000 reviews because of a large volume of unstructured data. With the help of Automatic Content Analysis, you can easily understand what is important to your customers.

Deep Learning can be used to Investigate the Digital Behaviour of consumers

A picture can tell thousands of words. Can we analyze images posted on social media and review platforms? Yes, we can. With the advancement of the Artificial Intelligence area, researchers can use deep learning to understand images. We have used deep learning to Investigate the Digital Behaviour of consumers.

In this research, 4,000 photos of nine Chinese restaurants posted on Tripadvisor’s website were analyzed using image recognition via Inception V3 and Google’s deep learning network; this revealed 12 hierarchical image clusters. Two examples of these clusters are provided below-

Cluster Name: Atmospheric cluster
This cluster of images is generated through the help of deep learning.

Cluster Name: Decorations using lamps in Restaurants
This cluster of images is generated through the help of deep learning.

The above clusters provided us an idea of what type contents consumers are posting on social media. These will also help restaurants to formulate different strategies such as improving the atmospheric environment and decorating using lamps. Competitors can also learn about their competitors by analyzing the images posted by consumers on online platforms.

To know more about this research, please read my following article-

Hasan, M.R.Abdunurova, A.Wang, W.Zheng, J. and Shams, S.M.R. (2021), “Using deep learning to investigate digital behavior in culinary tourism”, Journal of Place Management and Development, Vol. 14 No. 1, pp. 43-65. https://doi.org/10.1108/JPMD-03-2020-0022