Searchable Descriptions

Freetrade lowers the barrier to investing. There will be lots of first time investors. Freetrade should make it as easy as possible to find the company they’re interested in - it is not always obvious.

Searching for Primark on the Discover screen should bring up ABF. Primark is mentioned in ABF’s share screen description, but not on the discover screen (it should really be mentioned in both - Primark drives over half the revenue and profits of the entire group).

Searching for Guinness on the Discover screen should bring up Diageo for the same reason.

Search should take into account the Share screen Description. I’m assuming that right now, these descriptions are manually entered - what will happen with these when the Investment Platform happens?

A recommender system with named entity recognition, as an example. Though implementing an NLP system won’t be cheap or easy. It’s a constant process of understanding the issue at hand, writing algorithms, testing, and improving that never ends (modern day data science or just science).

In the long run, knowing the user’s risk profile may be helpful (again, all that machine learning stuff):

Check out this post about Robinhood and “Customers also bought”:

I believe that the absence of a good Discovery experience from the user journey on the product is an Achilles Heel for Robinhood. It opens Robinhood to disintermediation both by startups that might provide an intuitive Discovery platform and by incumbents who can build innovative products on top of their existing customer base. Erica, a digital financial assistant by Bank of America is an excellent example of that. Therefore, solving for the Discovery not only provides a good user experience but also addresses the potential existential threat.

Robinhood has been trying to address this problem and has launched many innovative features in the past few months: Collections, a revamped Search, and “People Also Bought.”


(Source - https://uxdesign.cc/robinhood-and-the-overkill-of-customers-also-bought-234266c525f)

But the case for purchasing stocks couldn’t be more different than buying products. The reason one person might be interested in a particular stock could be completely different from another person’s. And unlike a physical product, these motivations aren’t just a handful.

Let’s pick Amazon stock as an example. It could be of interest to a person who wants to add more growth stocks to the portfolio. It could be of interest to another person interested in increasing the weight of technology stocks or consumer discretionary stocks in the portfolio in her 401K. And another person might want to buy Amazon to hedge against the risk of other retailer stocks in the portfolio. And it goes on and on ad infinitum.

An ideal replacement of this flawed idea would be a feature that recommends stocks based on the risk profile, investment goal, current portfolio, desired portfolio allocation, etc. Something of the sort where Robinhood will look at my portfolio allocation to understand the kind of investor I’m and then recommends stocks.

1 Like

These posts by Aibnb’s Engineering team are quite good, their data science team is powers every department inside the company:

  1. High level overview of knowledge access and retrieval:

The knowledge graph is not a new concept. It has been used successfully at many companies (the most famous example being Google which uses it to power their search engine and surface relevant context for particular queries).

Normally, engineers work with relational structures, where a schema defines what each row of data contains. This is the preferred way for holding data because it works great for transactional processes since it makes it really quick to access rows of data. However, there is an operational burden when you have many table for distinct objects that may contain the same relational information in individual columns (ex: the city homes or experiences are located in, or the type of activity that an experience and that a destination is known for). This is where the graph structure comes into play.

Full post - https://medium.com/airbnb-engineering/scaling-knowledge-access-and-retrieval-at-airbnb-665b6ba21e95

  1. Same as above but more in-depth:

Imagine you are planning a trip to Los Angeles. The first step is to visit Airbnb.com and search for “Los Angeles.” On the backend, the query “Los Angeles” is translated into a block on the map; available Homes within this block are returned in many pages of search results. Is that enough for you to make your trip plan?

To scale our ability to answer these travel queries, we needed a systematic approach to storing and serving high-quality information about entities (e.g. cities, landmarks, events, etc.) and the relationships between them (e.g. the most popular landmark in a city, the best neighborhood for tacos, etc.). To tackle this problem, we have spent the past year building and applying a knowledge graph that stores and serves structured data that connects what makes our inventory unique, what our users are looking for, and what the world of travel has to offer.

Full post - https://medium.com/airbnb-engineering/contextualizing-airbnb-by-building-knowledge-graph-b7077e268d5a