Desktop version

Home arrow Computer Science arrow Applied Big Data Analytics in Operations Management

Fake Reviews, Leading to Opinion Frauds

In the recent times, it is being observed that a lot of fraud reviews are being created by fraudsters to distort the quality of a product. Such fake reviews can lead to the exponential decrease in the revenue of a firm if all the decisions of identifying the customers’ sentiments are wholly made using the sentiment analysis. Moreover, identification of such fraudulent reviews is an another tedious task which is still not been automated.

Figure 3. Cons of sentiment analysis in operation management

Shortage of Skilled Staff of NLP

The task of sentiment analysis covers machine learning and NLP techniques. Both these techniques are current and new which has not been catered by number of people in the recent time. Due to the shortage of skilled staff in this area, automation of social media posts is difficult in its own terms. A large number of organizations won’t be applying sentiment analysis in operation management due to the added overhead of training people in the new field which would add to the overall cost of a firm.

Threats from Illegitimate Data

A new term ‘opinion warfare’ is introduced in which firms create fraudulent reviews to mislead their competent firms. As in Figure 4, consider two firms-Firm A and Firm B competing in the market. It is being assumed that the Firm A takes its operation management decisions based on the sentiment analysis. Suppose Firm B gets to know about this and it deliberately create negative reviews about the product A of Firm A. Due to the negative comments about the product A, the Firm A will decrease the procurement of the Product A which will ultimately lead to the fall in the sales of the Firm A for product A. This type of situation can create a lot of confusion for the organizations and can lead to bad decision making. Detecting of such misleading posts can be a bit tricky as automated tools are required to identify such malicious attacks.

Figure 4. Firm B creates afalse-positive outcome that misdirects the efforts of Firm A (Wood et. al 2013a)

< Prev   CONTENTS   Source   Next >

Related topics