Using the correct techniques you can extract a lot of information from Ordnance Survey Mastemap Topo when it is combined with other data.
E.g. Show me all residential buildings over 18m tall (draw your own conclusions)
This method uses a database but it could be done with shape files and text files but a database have other advantages for further analysis.
To answer your question you will need the following:
Ordnance Survey MasterMap Topo for your selected area
Ordnance Survey AddressBase Plus for your area
The Addressbase Plus header file.
Load the Mastermap Topo data into a geodatabase. (You can use the Esri FGDB or another database such as Postgresql).
Load this data as a layer in ArcMap and perform a definition query Theme = ‘Buildings’. (This is not totally essential but it will make future steps a bit quicker)
Using whichever technique you prefer load the Addressbase Plus data into a table in your database. (My preference is either Pentaho or Talend; both have free versions)
Perform a join between Mastermap and Addressbase using Toid as the common field. At this point you have the choice of only keeping only the matched toads or all of them.
At this point you have a number of options. You could then perform a select query on class to find residential properties or one of the many address classifications included in Addressbase. You can also symbolise on code which may be much more useful as if you colour each class differently or use the field calculator to aggregate the classification into much broader classes (I usually do this at the load stage using Pentaho)
If you also download the building height data and join it using Toid you can then identify all the high-rise buildings in your area. Combined with Addressbase this lets you produce an address list of all these properties.
A minor caveat of this is that the height data was published in December 2014 and hasn’t been updated.
If you need more detailed classifications can I sugest taking a look at ukmap from Verisk http://www.geoinformationgroup.co.uk/ukmap which has much more detailed classifications than provided by Ordnance Survey, although for a much more limited coverage