Browse Prior Art Database

Cognitive incoming text screening cell phone app. Disclosure Number: IPCOM000253601D
Publication Date: 2018-Apr-16
Document File: 5 page(s) / 150K

Publishing Venue

The Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 37% of the total text.

Cognitive incoming text screening cell phone app.

If a mobile phone number is very similar to another phone number or local business number, then users often receive "mis-addressed" texts. In addition to becoming annoying for the recipient, mis-addressed texts can contain undesirable information (e.g., profanity, lewd attachments, unrelated personal or business data, etc.).

The novel contribution is a cognitive, incoming text-screening solution that examines the area code of incoming texts, text conversation durations, and frequencies of the incoming text conversations to accurately determine whether to prompt the user to receive a text. This capability provides more granularity in determining what to do with each local (i.e., within an area code), or non-local (i.e., outside an area code) incoming text. The novel system utilizes machine learning to build an incoming text number database (db), which can be queried later for faster and more accurate incoming text screening.

Analysis of an incoming phone text's in a local area code is useful in cases such as when a particular cell phone number is very close (i.e. one digit off) from a local business phone number or a re-used phone number from a previous phone user, or when a one or more dialed digits' physical location on the texter's dial pad are very close numerically to the unlucky recipient's number. In any of these cases, any unwanted incoming texts can be alleviated by the current system with one area code.

This system can screen texts coming in from any area code. The gist of the current cognitive approach, before prompting the user upon receiving an incoming text, is the ability to examine the area code of the incoming text first, then examine a db for text conversation duration calculated from any existing previous text conversations from the same number, and the frequency of any existing text conversations from the same number, to determine whether to screen the next time a text from the same incoming number is received. After a certain threshold is reached, the system learns which text numbers are mis-addressed, and then always screen the numbers for the user from that point forward. The system can also refer to a cell phone's blocked call list db to further assist in this effort.

Basic logic:

1. When a “within-area-code” text comes to the mobile phone, the system searches the app db to determine whether the number has at least one previous instance of a text conversation within a user-defined period

A. If the incoming text phone number does not exist in the db, then the system adds both the text phone number and duration of the calculated text conversation to the db

B. If the incoming number exists in the db, then the system adds only the duration of the text conversation to that incoming text phone number's db entry

2. If a within-area-code text number exists in the db AND it has at least one short- duration text (the latter is a configurable setting), then the system considers the...