On the Bs'rii Language (Formerly the Urmer)

SSA has been a passion of mine for about 2 years now, so I figured hey why not! I've also included a sketch of the Bs'r.

^Abstract. ^Subjectivity ^and ^sentiment ^analysis ^has ^recently ^gained ^considerable ^attention ^amongst ^scholars ^of ^the ^Imperial ^Linguistics ^Society, ^but ^most ^of ^the ^resources ^built ^so ^far ^are ^tailored ^to ^Cyrodiilic ^and ^other ^local ^languages. ^The ^need ^for ^designing ^systems ^for ^other ^languages ^is ^increasing, ^especially ^as ^trade ^and ^cross-community ^intermingling ^become ^popular ^throughout ^our ^world. ^This ^paper ^surveys ^different ^techniques ^for ^SSA ^for ^the ^Bs’rii ^language. ^After ^a ^brief ^synopsis ^about ^Bs’rii, ^I ^describe ^the ^main ^existing ^techniques ^and ^test ^corpora ^that ^have ^been ^introduced ^in ^the ^literature.

The language of the elusive Bs’r is a difficult subject to write about. Spoken by an estimated five thousand people it boasts a small population with no speakers outside natives. It consists of multiple dialects that differ across regions, with each qualifying almost indefinitely as their own languages. There is no standard across communities, and from what I’ve seen the general speaking population does not want there to be. I’ve been lucky enough to live for some weeks amongst a tribe in the Wrothgarian mountains. What I have translated is broken, and confusing to me— what I haven’t, I’m afraid I never will. What isn’t a click or tap is an inaudible whine, detected only by other Bs’r, using specialized organs located in the chest. I can never tell when there’s been a break in the conversation, and when I’m being spoken to in a dialect out of my reach. Regardless, I’ve properly record that which I could glean from my gracious hosts. This already makes for difficulties in recording and decoding the language using traditional methods.

A quick and raspy chirp, which I have designated as their R, is indicative of nouns and proper subjects when used as a suffix— thusly their name, Bs’r, is something akin to “connected folk” in Tamrielic. The title speaks to their close-knit colonies.

There are few vowels; they have no need for them. Chirps and clicks make up the vast majority of the language, and I have a hard time replicating them— however there are some similar sounds. Particularly T, B and S appear as they would in my own native tongue, and are produced in much a similar fashion. Semantics are dependent on phonetics here, which each individual sound providing a part of speech. There are very, very few individual words I’ve found— sounds are repeated at different points to indicate different things, instead. For example, two Ts followed by an R indicates a living thing, while one T followed by an R, and a final T, indicates multiple living things, or people. As such, the phrase “the legions are coming over the hills,” can be translated directly as “many people move distance.” It’s fascinating really, what they’ve accomplished with so little. It makes me wonder about the origins of Orcish and its old ties.

One way of building sentiment analyses for languages other than Cyrodiilic or building systems that work for multiple languages is to extract and select features that do not depend on the language itself. Different approaches have been followed to select and extract these features, including: (1) Entropy Weighted Genetic algorithms, (2) Feature Subsumption, (3) Local grammar-based methods, (4) Positional features and (5) Common seeds word methods.

All have failed when applied to Bs’rii.