The sentiment score in Sentalyse captures how a company or its components are perceived over time. Rather than reacting to single news events, it reflects a cumulative picture based on sustained tone in reporting. This long-term view ensures that one-off positive articles don’t immediately outweigh a longer trend of negative sentiment. It helps analysts focus on structural reputation signals, not daily noise.

Sentalyse primarily gathers data from reliable news platforms and respected financial communities. We aim to ensure source credibility and consistency by filtering out untrustworthy or unverified content, so your sentiment analysis is based on facts and professional reporting.

We use high-quality, publicly available data from reputable news sources. Articles are processed and validated through our data pipeline to remove duplicates, spam, and irrelevant content. This ensures that the sentiment calculations are grounded in accurate and meaningful reporting.

Our sentiment data is updated on a regular basis, with new articles being collected, analyzed, and integrated into the overall sentiment score. This keeps your insights up to date with market developments while preserving the long-term memory of previous trends.

Sentalyse tracks sentiment across a comprehensive range of components that reflect a company’s internal and external performance. These include: Company, Technology, Sustainability, HR, Operations, Marketing, Management, Shareholders, Legal, Financials, Macroeconomic Exposure, Customer & Public Sentiment, and Strategy & Innovation. Each component is measured individually to allow for granular analysis.

Long-term sentiment values range from -1 to 1. A score of -1 indicates sustained negative sentiment, 0 represents a neutral perception, and 1 reflects consistently positive sentiment. These values are not just momentary signals, but reflect the emotional trend over an extended period.

Many news articles contain both optimistic and pessimistic statements. For example, a report may highlight a company's innovation (positive) while criticizing its labor practices (negative). Sentalyse captures these mixed tones by allowing both positive and negative sentiment to be recorded for the same article.

Sentiment is considered strong when the score approaches -1 or 1, which indicates a clear, consistent tone in reporting. Scores near 0 suggest a more balanced or uncertain perception. Strong sentiment values are especially useful in highlighting risk or opportunity.

Sentalyse uses advanced Natural Language Processing (NLP) models based on deep learning architectures. These models are trained to detect sentiment with high contextual accuracy, going beyond keyword spotting to understand nuance, tone, and topic relevance.

Each component is mapped to a set of related keywords and phrases. For instance, the component 'Sustainability' is associated with terms like 'renewable', 'carbon-neutral', and 'circularity'. The sentiment of articles mentioning these terms in context is attributed to that component, creating a specific sentiment profile for each area of the business.

Long-term sentiment is calculated using proprietary smoothing methods that mimic cognitive psychology principles, such as memory decay and relevance over time. Older articles still influence sentiment but with diminishing weight, reflecting how perceptions evolve in real life. This method balances recency with reputation history.

Long-term sentiment is useful for a range of professionals: - Investors: to detect early signs of reputational risk or opportunity. - Financial analysts: to enrich fundamental analysis with narrative trends. - Traders: to incorporate media signals into timing strategies. - Due diligence teams: to assess ESG performance or red flags before acquisition. It’s particularly valuable when used alongside traditional financial metrics.