Sentiment

The goal of sentiment analysis is to automatically classify text data as positive, negative, or neutral based on the underlying sentiment expressed in the text.

To achieve this Lettria analyzes various linguistic features of the text, such as word choice, syntax, and sentiment-laden words or phrases.

These features are then fed into an algorithm, such as a machine learning model, which is trained to recognize patterns and trends in the data that are indicative of sentiment.

One approach to text sentiment analysis combines the strengths of rule-based algorithms that you can set up as part of your solution on the Lettria platform, with machine learning algorithms that are pre-trained, as well as a solution that is built on the platform with your own data and annotations.

This helps improve the accuracy and efficiency of the sentiment analysis process, as well as reduce the need for manual labeling or annotation of data.

{
  "start_id": 0,
  "end_id": 2,
  "elements": [
    {
      "target": null,
      "subject": {
        "lemma": "Victor",
        "source": "Victor",
        "index": 0
      },
      "value": 0.27,
      "source": { "index": 1, "lemma": "etre", "source": "est" }
    }
  ],
  "sentence": "Victor est la",
  "values" : { "positive": 0.454, "negative": 0, "total": 0.454 }
}

{
	"target": null,
	"subject": { "lemma": "Victor", "source": "Victor", "index": 0 },
	"value": 0.27,
	"source": { "index": 1, "lemma": "etre", "source": "est" }
}

{ "positive": 0.27, "negative": 0, "total": 0.27 }

{ "lemma": "immeuble", "source": "immeubles", "index": 6 }

{ "lemma": "Victor", "source": "Victor", "index": 0 }

• Emotion