Why we don’t use personal account tracing on social media
Using event extraction from a personal account on Twitter can be a risky method for news sentiment assessment. This week, we got a very good example of why we don’t cover personal twitter account on our platform.
On Aug 07, Tesla CEO Alon Musk tweets he is ‘considering’ taking the company private, investor react immediately and increase and entering the stock which sends the share price up. For companies that monitoring Along Musk account, this event is a very bullish one, but the first red flag was the distribution of the news as SEC and regulation forbade distribution key events via other platforms rather than the regulated one. therefore, the outcome was inevitable, after 2 days the gaining from this event was gone as for the investors that jump in on Aug 07 it is a sure loss.
A personal account on twitter or other social media holds a tremendus risk of misleading event detection. First, it is not allowed to share important news and events about public companies outside the regulated filling method. Second, a personal account can be hacked and could be used in the fake news which will cause a damage for investors that relly on news.finally, personal accounts could hold a non conformed event which can effect the sentiment model when evaluating new events from the news.
Using twitter monitoring as a news events source can be beneficial, at FIRST TO INVEST we use corporate and financial publication account the see and analyze the news and event flow and usually you can get a fast detection of a new event after the proper filling is made and before the financial publication report on them:
“FT Exclusive: Saudi Arabia’s sovereign wealth fund has built a $2bn stake in Tesla https://on.ft.com/2MaSSB1”
“Tesla starts hiring for new $2 billion Shanghai plant https://yhoo.it/2ngsoA1 pic.twitter.com/MFhzyQ6VFR.”
To conclude, the next time your firm evaluate a data provider that cover social media, ask about the methodology of evaluating events for social media in order to reduce the risk and losses
(Event-driven score is scaled between 0-1 where bellow 0.48 consider as bearish based on events during the filtered period, above 0.52 consider bullish)
Using big data and NLP technologies to capture alpha by collecting, structuring, and revealing events from news articles, press releases, and financial social media.
(Views and recommendations given in this section are for research purposes only. Please consult your financial adviser before taking any position in the stock/s or currencies mentioned.) Neither First to invest. nor any of its officers, employees, representatives, agents or independent contractors are, in such capacities, licensed financial advisors, registered investment advisers or registered broker-dealers. First to invest does not provide investment or financial advice or make investment recommendations. Nothing contained in this communication constitutes a solicitation, recommendation, promotion, endorsement or offer by First to invest of any particular security, transaction or investment.)