Women Communicate Better: Classy

Everyone’s familiar with a class-action lawsuit where a bunch of families get together to sue a pharmaceutical company. Well, class-action lawsuits can be filed by shareholders, too. When a company acts fraudulently by issuing misleading statements to investors or hiding negative information about their firm, they can cause economic injury to shareholders (since stock prices will generally drop upon eventual disclosure of the information.) When that happens, the primary means of recourse is through the legal system. These cases are called securities class-action lawsuits.

The Problem:

Currently, there are a handful of firms that specialize in this type of litigation. These plaintiff firms basically throw bodies at the problem by keeping tons of lawyers on staff. These employees manually comb through the news, read reports from industry bloggers, and keep their eye on the stock exchange, hoping that they can identify a possible opportunity for a suit before one of their competitors. This is an incredibly time-intensive and inefficient process. If a suit isn’t identified immediately, it takes an average of 77 days for a plaintiff firm to file.

Since law firms invest so much manpower in the identification process, they are incentivized to only take on the largest companies with the highest potential settlements. The average market capitalization for firms targeted by securities class action suits in 2016 was $9.08 Billion. This means that plaintiff firms are neglecting to hold the majority of malfeasant companies accountable for their actions because either they’re not big enough fish or lawyers are simply missing these opportunities because fraud in smaller companies doesn’t dominate the news cycle.

These firms generally take a shotgun approach to bringing these securities suits to court, leaning on high frequency and volume over quality. Historically, 44% of all securities class-action suits are dismissed, meaning that the courts are being inundated with frivolous cases. And it’s increasing year over year:

*Source: Stanford Law School

The Classy Solution:

Instead of relying on plaintiff lawyers and industry blogs (like Lyle Roger’s The 10b-5 Daily) to manually scan and analyze stock price data, we believe there is an opportunity to use machine learning to drive a uniquely efficient, highly competitive plaintiff firm that can hold more corporations accountable for their actions.

We propose the creation of Classy, a revolutionary algorithm that utilizes machine learning and human input to classify and predict securities lawsuits.

Our value proposition is threefold:

1. Bring Suits Faster

The Classy algorithm will be able to identify potential suits much more efficiently than traditional means. Filing the strongest, most profitable suits before a rival firm gives our firm a huge competitive edge.

The more efficient the system, the faster we can hold companies accountable for their actions and address client grievances.

2. Cover the Whole Market

Classy allow the plaintiff firm to hold companies of all sizes accountable, not just the largest, most profitable ones. Using augmented perception, we can analyze the entire market for stock patterns that demonstrate the potential for a suit (something that is impossible to do manually).

The algorithm can also cover a much larger breadth of news sources than a human analyst. This gives us a better ability to match a news item with a companion drop in stock prices that captures the potential for fiduciary misconduct.

3. Choose Better Cases

Classy can reduce the volume of frivolous lawsuits that get filed by allowing the firm to better prioritize their staffing structure and redistribute their human and financial resources away from searching for potential fraud and towards suit selection and execution.

Once the algorithm has enough data, we can extend its functionality to issue a prediction score to the firm, which would encompass the likelihood of winning the case as well as the estimated settlement value. The plaintiff firm can use this tool to help guide them away from filing unsuccessful, unproductive cases.

A Classy Design:

Classy combines external sensors with machine and human algorithms to predict the likelihood of securities misconduct of various firms and help analyze the success of a suit.

1. Monitor stock prices (sensory input)

We would use supervised machine learning and deep learning to flag precipitous stock price drops throughout the whole market.

2. Track relevant sentiment (sensory input)

We would use natural language processing and sentiment analysis to analyze relevant news items, identifying patterns of negative disclosures by a firm in the past or public apologies issued by CEOs.

3. Predict Outcomes (machine algorithm)

These sensory inputs would then be analyzed by a machine algorithm, which would use the data to create a likelihood score of disclosure malfeasance by the firm and the predicted settlement value.

4. Supervise with experienced plaintiff attorneys (human algorithm)

This information would then be transmitted through a human algorithm – plaintiff lawyers with years of experience and relationship expertise – who would then verify and expand upon the potential suits flagged by the machine algorithm. They would also provide feedback to the machine algorithm in order to improve its efficacy and accuracy over time.

Action Steps:

Validation: We want to assess our hypothesis by using historical data to test the validity of our claims: that Classy can bring cases faster, with a lower margin of error, and extend to a broader set of companies.

Poach a Partner: Rather than offer Classy as an available service to all plaintiff firms (which could lead to a race to the bottom), we would like to partner with a set of experienced plaintiff lawyers and start our own firm. This will give us a competitive edge over rival firms and a greater potential to monetize our efforts.

Start Bringing Suits!

Our Funding Target:

We believe that validation can be achieved in two months and we have budgeted accordingly: