{"id":19856,"date":"2020-01-28T14:34:42","date_gmt":"2020-01-28T14:34:42","guid":{"rendered":"https:\/\/www.arimetrics.com\/glosario-digital\/sentiment-analysis"},"modified":"2026-05-11T23:02:30","modified_gmt":"2026-05-11T23:02:30","slug":"sentiment-analysis","status":"publish","type":"encyclopedia","link":"https:\/\/www.arimetrics.com\/en\/digital-glossary\/sentiment-analysis","title":{"rendered":"Sentiment Analysis"},"content":{"rendered":"

\"Sentiment<\/p>\n

Definition: <\/strong><\/p>\n

Sentiment analysis <\/em>is a type of natural language processing to track the public’s mood about a particular product.<\/p>\n

Sentiment analysis involves building a system for collecting and categorizing opinions about a product. Also known as opinion mining, it can be automated using machine learning, a type of artificial intelligence (AI), to extract sentiment from text.<\/p>\n

 <\/p>\n\n

Applications of sentiment analysis<\/h2>\n

With this system, you can help marketers evaluate the success of an advertising campaign<\/a> or new product launch or analyze whether versions of a product or service are popular.<\/p>\n

For example, a review from a website might be quite positive about a digital camera, but at the same time it is specifically negative about how heavy this camera is. Being able to identify this type of information in a systematic way gives the provider a much clearer picture about public opinion than surveys or focus groups<\/a>, because the data is provided by the customer directly.<\/p>\n

How sentiment analysis works<\/h2>\n
    \n
  • We analyze the exact intent of a phrase to know if it speaks of a brand or a product.
    \nWe value the phrase through polarity, which classifies the message according to the author’s intention, and can be positive, negative or neutral. With this you control the feeling of the users.<\/li>\n
  • Processing of the information of the previous points. This process is divided into:\n
      \n
    • Manual: if the word or expression can have different meanings depending on the field, that word will have to be classified manually where appropriate.<\/li>\n
    • Automatic analysis: different keywords are established that frame the expressions depending on what words or set of similar words you have. An example of this could be that any phrase that carried “good”, “beautiful”, “beautiful” was included in the positive feeling section.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n

      Limitations of Automatic Sentiment Analysis<\/h2>\n

      The main one is that it is not as effective in ranking as manual sentiment analysis can be.<\/p>\n

      In addition, the programs that are responsible for the automation of these tasks are not able to differentiate exactly if a sentence contains sarcasm or irony and, therefore, could confuse a sarcastic phrase of the style “yes, yes, that product (x) is very good but on the other end”. Therefore, the vast majority of software are not able to know the real intention of a sentence and, consequently, catalog it as positive or negative.<\/p>\n

      There are certain programs that are able to differentiate a sincere phrase from a more ironic one but they are somewhat more complex to configure.<\/p>\n

      Frequently asked questions about Sentiment Analysis<\/h2>\n
      \n
      \nWhat does Sentiment Analysis mean in digital marketing?<\/summary>\n

      Sentiment Analysis refers to the concept described in this glossary entry: Definition: Sentiment analysis is a type of natural language processing to track the public's mood about a particular product. Sentiment analysis involves building a system for collecting and categorizing opinions about a product. It gives teams a shared vocabulary for analysing digital projects.<\/p>\n<\/details>\n

      \nWhen should teams pay attention to Sentiment Analysis?<\/summary>\n

      Teams should review Sentiment Analysis when it affects acquisition, measurement, user experience, content, automation or campaign performance. The important step is to connect the definition with a real decision.<\/p>\n<\/details>\n

      \nHow is Sentiment Analysis used in a digital strategy?<\/summary>\n

      Sentiment Analysis is used by translating the concept into practical checks: where it appears in the funnel, which data or channel is involved and whether it needs optimisation, monitoring or documentation.<\/p>\n<\/details>\n

      \nWhat is a common mistake when interpreting Sentiment Analysis?<\/summary>\n

      A common mistake is using Sentiment Analysis too broadly. It is better to verify the context, the tool or the metric involved before making strategic or technical conclusions.<\/p>\n<\/details>\n<\/div>\n