Semantic analysis is the study of the meaning of language, whereas sentiment analysis represents the emotional value. But what do those terms mean specifically?
In the field of content analysis the challenge is to capture high-level semantics. Machines can not process this task as naturally as humans. Before a machine is able to handle semantics automatically, a heavily time-consuming manual labeling process is needed, going hand-in-hand with data-learning.
At the same time, we have to avoid redundancy, i.e. the same content distributed several times on several channels.
A big amount of information causes limited control over the received content; therefore TV content analysis is a core requirement.
Subjects of news stories are usually persons, locations, and organisations. Identifying the names of these entities in the text stream allows us to do a first-level semantic analysis and gives us a powerful means of focusing our search results.
Like in audio segmentation, a continuous stream becomes better manageable once it can be split up into segments. On textual data we can use a semantic analysis tool to find boundaries between stories in the text. A story is defined as a region of text which treats a particular topic. Topic detection and story segmentation work in combination to define stories and give those stories a semantic label called topic.
Emotions are present in every social interaction, and in all forms of communication; they express opinions and information shared in the media. Therefore, emotions play a key role in industry, business decisions, marketing, sales, and define business success. This indicates that extracting emotional value provides crucial information for decision makers.
Sentiment also means the “emotional” part/content of a sentence or the whole document.
Applying a novel technology, the sentiment analysis, we can classify the polarity from various types of media sources (TV, radio, newspapers, online sources). The polarity is defined as “Whether the expressed opinion in a document, a sentence or an entity feature/aspect is positive, negative, or neutral. (= the polarity)" (Source of definition: wikipedia.org). Organizations can benefit from Sentiment Analysis to determine what is being said – positive, negative, neutral – about the organization itself, key events, key persons, its brands, and measure the impact of a press release, for example.
Find more information about the other technologies used by eMM: