Search Results

Search found 32453 results on 1299 pages for 'acts as list'.

Page 3/1299 | < Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >

  • List comprehension from multiple sources in Python?

    - by Noah
    Is it possible to replace the following with a list comprehension? res = [] for a, _, c in myList: for i in c: res.append((a, i)) For example: # Input myList = [("Foo", None, [1, 2, 3]), ("Bar", None, ["i", "j"])] # Output res = [("Foo", 1), ("Foo", 2), ("Foo", 3), ("Bar", "i"), ("Bar", "j")]

    Read the article

  • How do I process a nested list?

    - by ddbeck
    Suppose I have a bulleted list like this: * list item 1 * list item 2 (a parent) ** list item 3 (a child of list item 2) ** list item 4 (a child of list item 2 as well) *** list item 5 (a child of list item 4 and a grand-child of list item 2) * list item 6 I'd like to parse that into a nested list or some other data structure which makes the parent-child relationship between elements explicit (rather than depending on their contents and relative position). For example, here's a list of tuples containing an item and a list of its children (and so forth): [('list item 1',), ('list item 2', [('list item 3',), [('list item 4', [('list item 5'),]] ('list item 6',)] I've attempted to do this with plain Python and some experimentation with Pyparsing, but I'm not making progress. I'm left with two major questions: What's the strategy I need to employ to make this work? I know recursion is part of the solution, but I'm having a hard time making the connection between this and, say, a Fibonacci sequence. I'm certain I'm not the first person to have done this, but I don't know the terminology of the problem to make fruitful searches for more information on this topic. What problems are related to this so that I can learn more about solving these kinds of problems in general?

    Read the article

  • Haskell - function (that returns a list) on each element in a list

    - by Ben
    The assignment is to create a multiples function and I essentially want todo the following code: map (\t -> scanl (\x y -> x+y) t (repeat t)) listofnumbers The problem is that the scanl function returns a list of results rather than the one which the map function requires. So is there a function that will allow the return of lists?

    Read the article

  • List Component that acts as if control was permanently pressed

    - by Dennkster
    Hey, I have a list control and i want the user to be able to select many items at a time. Thus I want it to act that if the control key is pressed while he is clicking. Eg if he clicks on a selected row it should become unselected and if he clicks on a unselected row it should become selected. Do you have any idea how to do this? Thanks, Dennis

    Read the article

  • Random List of numbers in C

    - by Ant
    I have just started a C programming course and so far have only done the basics like printf, read a little on variables etc on the course book. The teacher has tasked us with writing a program that will take a value entered by the user, this will be the number of students in the class (25 at the moment but can be variable). It will then list the number of students randomly and place them in 3 columns. The purpose is to sort students into groups of 3 randomly, then display on the screen in columns. Now my question is not for the code that defeats the object of me attempting the exercise, but how to structure it. I can see that using an array can be used to display the list, but really after some pointers on how best to approach the problem in blocks, then I can attempt to program each block.

    Read the article

  • trying to append a list, but something breaks

    - by romunov
    I'm trying to create an empty list which will have as many elements as there are num.of.walkers. I then try to append, to each created element, a new sub-list (length of new sub-list corresponds to a value in a. When I fiddle around in R everything goes smooth: list.of.dist[[1]] <- vector("list", a[1]) list.of.dist[[2]] <- vector("list", a[2]) list.of.dist[[3]] <- vector("list", a[3]) list.of.dist[[4]] <- vector("list", a[4]) I then try to write a function. Here is my feeble attempt that results in an error. Can someone chip in what am I doing wrong? countNumberOfWalks <- function(walk.df) { list.of.walkers <- sort(unique(walk.df$label)) num.of.walkers <- length(unique(walk.df$label)) #Pre-allocate objects for further manipulation list.of.dist <- vector("list", num.of.walkers) a <- c() # Count the number of walks per walker. for (i in list.of.walkers) { a[i] <- nrow(walk.df[walk.df$label == i,]) } a <- as.vector(a) # Add a sublist (length = number of walks) for each walker. for (i in i:num.of.walkers) { list.of.dist[[i]] <- vector("list", a[i]) } return(list.of.dist) } > num.of.walks.per.walker <- countNumberOfWalks(walk.df) Error in vector("list", a[i]) : vector size cannot be NA

    Read the article

  • Using FindAll on a List<List<T>> type

    - by Ken Foster
    Assuming public class MyClass { public int ID {get; set; } public string Name {get; set; } } and List<MyClass> classList = //populate with MyClass instances of various IDs I can do List<MyClass> result = classList.FindAll(class => class.ID == 123); and that will give me a list of just classes with ID = 123. Works great, looks elegant. Now, if I had List<List<MyClass>> listOfClassLists = //populate with Lists of MyClass instances How do I get a filtered list where the lists themselves are filtered. I tried List<List<MyClass>> result = listOfClassLists.FindAll (list => list.FindAll(class => class.ID == 123).Count > 0); it looks elegant, but doesn't work. It only includes Lists of classes where at least one class has an ID of 123, but it includes ALL MyClass instances in that list, not just the ones that match. I ended up having to do List<List<MyClass>> result = Results(listOfClassLists, 123); private List<List<MyClass>> Results(List<List<MyClass>> myListOfLists, int id) { List<List<MyClass>> results = new List<List<MyClass>>(); foreach (List<MyClass> myClassList in myListOfLists) { List<MyClass> subList = myClassList.FindAll(myClass => myClass.ID == id); if (subList.Count > 0) results.Add(subList); } return results; } which gets the job done, but isn't that elegant. Just looking for better ways to do a FindAll on a List of Lists. Ken

    Read the article

  • Unique elements of list within list in python

    - by user2901061
    We are given a list of animals in different zoos and need to find which zoos have animals that are not in any others. The animals of each zoo are separated by spaces, and each zoo is originally separated by a comma. I am currently enumerating over all of the zoos to split each animal and create lists within lists for different zoos as such: for i, zoo in enumerate(zoos): zoos[i] = zoo.split() However, I then do not know how to tell and count how many of the zoos have unique animals. I figure it is something else with enumerate and possibly sets, but cannot get it down exactly. Any help is greatly appreciated. Thanks

    Read the article

  • Optimizing list comprehension to find pairs of co-prime numbers

    - by user3685422
    Given A,B print the number of pairs (a,b) such that GCD(a,b)=1 and 1<=a<=A and 1<=b<=B. Here is my answer: return len([(x,y) for x in range(1,A+1) for y in range(1,B+1) if gcd(x,y) == 1]) My answer works fine for small ranges but takes enough time if the range is increased. such as 1 <= A <= 10^5 1 <= B <= 10^5 is there a better way to write this or can this be optimized?

    Read the article

  • Counting number of times an item occurs in a linked list

    - by HanaCHaN92
    Here is the assignment: Here's the assignment: Implement a method countValue() that counts the number of times an item occurs in a linked list. Remember to use the STL list. int countValue(list front, const int item); Generate 20 random numbers in the range of 0 to 4, and insert each number in the linked list. Output the list by using a method which you would call writeLinkedList which you would add to the ListP.cpp. In a loop, call the method countValue() , and display the number of occurrences of each value from 0 to 4 in the list. Remember that all the above is to be included in the file ListP.ccp Run: 2 3 4 0 1 0 2 4 2 3 3 4 3 3 3 0 0 2 0 2 0 : 5, 1 : 1, 2 : 5, 3 : 6, 4 : 3 and here is what I have so far: #include<iostream> #include<list> #include<tchar.h> int countValue(list<int> front, const int item); using namespace std; int _tmain(int argc, _TCHAR* argv[]){ list<int> front; int listCount; cout << "Enter the size of the list: "; cin >> listCount; for (int i = 1; i <= listCount; i++) front.insert(rand()%5); cout << "Original List of Values: " << endl; //writeLinkedList(front, " "); cout << endl; for(int j=0;j<5;++j) cout << countValue (front,j) << endl; cout << endl; return 0; } int countValue(list<int> front, const int item) { int count0; int count1; int count2; int count3; int count4; list<int> *List; for(list<int>::iterator i = front.begin(); i != front.end(); i++) { if(List->item == 0) { count0++; } if(List->item == 1) { count1++; } if(List->item == 2) { count2++; } if(List->item == 3) { count2++; }if(List->item == 4) { count4++; } } } And here are the errors: error C2065: 'list' : undeclared identifier line 5 error C2062: type 'int' unexpected line 5 error C2661: 'std::list<_Ty>::insert' : no overloaded function takes 1 arguments line 16 error C3861: 'countValue': identifier not found line 21 IntelliSense: no instance of overloaded function "std::list<_Ty, _Ax>::insert [with _Ty=int, _Ax=std::allocator<int>]" matches the argument list line 16 IntelliSense: too few arguments in function call line 16 error C2039: 'item': is not a member of 'std::list<_Ty>' lines 34, 38, 42, 46, 49 IntelliSense: declaration is incompatible with "int countValue" (declared at line 5) line 25 IntelliSense: class "std::list<int, std:: allocator<int>>" has no member "item" lines 34, 38, 42, 46, 49 I just want to know what I've done wrong and how to fix it and also if someone could help me figure out if I'm doing the countValue function wrong or not based on the instructions I would really appreciate it. I've read the chapter in our textbook several times, looked up tutorials on youtube and on Dream in Code, and still I can not figure this out. All helpful information is appreciated!

    Read the article

  • Best way to choose random element from weighted list

    - by Qqwy
    I want to create a simple game. Every so often, a power up should appear. Right now the different kinds of power ups are stored in an array. However, not every power up should appear equally often: For instance, a score multiplier should appear much more often than an extra life. What is the best/fastest way to pick an element at random from a list where some of the elements should be picked more often than others?

    Read the article

  • Joystick acts as mouse; won't stop

    - by Shazzner
    Joystick acts as a mouse, even when I'm playing a game that uses a joystick so I get random mouse events going on. I plugged a joystick in to play Spiral Knights, also installed joystick and jcalibrate. Everything is working good, except by default the joystick moves the mouse around and the button activate mouse keys. Now normally this would be good behavior if I'm on a Myth-box or something, unfortunately when I play Spiral Knights with joystick input I see my mouse cursor moving in the back ground and when I hit a button it thinks I'm pressing right-click so it minimizes everything. Also it creates folders and probably deletes stuff. So, basically how the heck do I stop it from acting as a mouse?

    Read the article

  • Search one element of a list in another list recursively

    - by androidnoob
    I have 2 lists old_name_list = [a-1234, a-1235, a-1236] new_name_list = [(a-1235, a-5321), (a-1236, a-6321), (a-1234, a-4321), ... ] I want to search recursively if the elements in old_name_list exist in new_name_list and returns the associated value with it, for eg. the first element in old_name_list returns a-4321, second element returns a-5321, and so on until old_name_list finishes. I have tried the following and it doesn't work for old_name, new_name in zip(old_name_list, new_name_list): if old_name in new_name[0]: print new_name[1] Is the method I am doing wrong or I have to make some minor changes to it? Thank you in advance.

    Read the article

  • setting up linked list Java

    - by erp
    I'm working on some basic linked list stuff, like insert, delete, go to the front or end of the list, and basically i understand the concept of all of that stuff once i have the list i guess but im having trouble setting up the list. I was wondering of you guys could tell me if im going in the right direction. (mostly just the setup) this is what i have so far: public class List { private int size; private List linkedList; List head; List cur; List next; /** * Creates an empty list. * @pre * @post */ public List(){ linkedList = new List(); this.head = null; cur = head; } /** * Delete the current element from this list. The element after the deleted element becomes the new current. * If that's not possible, then the element before the deleted element becomes the new current. * If that is also not possible, then you need to recognize what state the list is in and define current accordingly. * Nothing should be done if a delete is not possible. * @pre * @post */ public void delete(){ // delete size--; } /** * Get the value of the current element. If this is not possible, throw an IllegalArgumentException. * @pre the list is not empty * @post * @return value of the current element. */ public char get(){ return getItem(cur); } /** * Go to the last element of the list. If this is not possible, don't change the cursor. * @pre * @post */ public void goLast(){ while (cur.next != null){ cur = cur.next; } } /** * Advance the cursor to the next element. If this is not possible, don't change the cursor. * @pre * @post */ public void goNext(){ if(cur.next != null){ cur = cur.next;} //else do nothing } /** * Retreat the cursor to the previous element. If this is not possible, don't change the cursor. * @pre * @post */ public void goPrev(){ } /** * Go to top of the list. This is the position before the first element. * @pre * @post */ public void goTop(){ } /** * Go to first element of the list. If this is not possible, don't change the cursor. * @pre * @post */ public void goFirst(){ } /** * Insert the given parameter after the current element. The newly inserted element becomes the current element. * @pre * @post * @param newVal : value to insert after the current element. */ public void insert(char newVal){ cur.setItem(newVal); size++; } /** * Determines if this list is empty. Empty means this list has no elements. * @pre * @post * @return true if the list is empty. */ public boolean isEmpty(){ return head == null; } /** * Determines the size of the list. The size of the list is the number of elements in the list. * @pre * @post * @return size which is the number of elements in the list. */ public int size(){ return size; } public class Node { private char item; private Node next; public Node() { } public Node(char item) { this.item = item; } public Node(char item, Node next) { this.item = item; this.next = next; } public char getItem() { return this.item; } public void setItem(char item) { this.item = item; } public Node getNext() { return this.next; } public void setNext(Node next) { this.next = next; } } } I got the node class alright (well i think it works alright), but is it necessary to even have that class? or can i go about it without even using it (just curious). And for example on the method get() in the list class can i not call that getItem() method from the node class because it's getting an error even though i thought that was the whole point for the node class. bottom line i just wanna make sure im setting up the list right. Thanks for any help guys, im new to linked list's so bear with me!

    Read the article

  • What You Said: What’s on Your Geeky Christmas List

    - by Jason Fitzpatrick
    Earlier this week we asked you to share what’s on your geeky Christmas list; you responded and we’re back to share your longed for tech goodies. The most requested item was this year’s hot introduction to the project board market: the Raspberry Pi. Dave writes: A Rapsberry Pi to tinker with, especially to see if I can get it up and running with OpenElec/Raspbmc and a torrent client for a low power media centre/htpc We just finished setting up a batch of new 512MB Raspberry Pi systems running the newest release of Rasbmbc and can’t recommend it enough–new refinements in Raspbmc and the extra 256MB of RAM really improve the media center experience. All John wants is a real keyboard so he can escape the torture of using a touch screen: How to Factory Reset Your Android Phone or Tablet When It Won’t Boot Our Geek Trivia App for Windows 8 is Now Available Everywhere How To Boot Your Android Phone or Tablet Into Safe Mode

    Read the article

  • Cast Object to Generic List

    - by CrazyJoe
    I have 3 generict type list. List<Contact> = new List<Contact>(); List<Address> = new List<Address>(); List<Document> = new List<Document>(); And save it on a variable with type object. Now i nedd do Cast Back to List to perfom a foreach, some like this: List<Contact> = (List<Contact>)obj; But obj content change every time, and i have some like this: List<???> = (List<???>)obj; I have another variable holding current obj Type: Type t = typeof(obj); Can i do some thing like that??: List<t> = (List<t>)obj; Obs: I no the current type in the list but i need to cast , and i dont now another form instead: List<Contact> = new List<Contact>(); Help Plz!!!

    Read the article

  • Filter any mailing list in GMail using the "list:" meta-data

    - by Binary255
    Hi, If I ask GMail to create a filter for a mailing list it creates a rule containing list:mailing-list-identifier, in the case of the NAnt mailing list it wrote: Has the words: list: "nant-users.lists.sourceforge.net" Is there a way to filter any mailing list? I would like to filter conversations from any mailing list containing answers to things I've previously asked (or answered to). Part of that filter is identifying "anything which is part of a mailing list" and I'm wondering if there is a better way than adding another label to all mailing list posts (which is cumbersome).

    Read the article

  • SharePoint logging to a list

    - by Norgean
    I recently worked in an environment with several servers. Locating the correct SharePoint log file for error messages, or development trace calls, is cumbersome. And once the solution hit the cloud, it got even worse, as we had no access to the log files at all. Obviously we are not the only ones with this problem, and the current trend seems to be to log to a list. This had become an off-hour project, so rather than do the sensible thing and find a ready-made solution, I decided to do it the hard way. So! Fire up Visual Studio, create yet another empty SharePoint solution, and start to think of some requirements. Easy on/offI want to be able to turn list-logging on and off.Easy loggingFor me, this means being able to use string.Format.Easy filteringLet's have the possibility to add some filtering columns; category and severity, where severity can be "verbose", "warning" or "error". Easy on/off Well, that's easy. Create a new web feature. Add an event receiver, and create the list on activation of the feature. Tear the list down on de-activation. I chose not to create a new content type; I did not feel that it would give me anything extra. I based the list on the generic list - I think a better choice would have been the announcement type. Approximately: public void CreateLog(SPWeb web)         {             var list = web.Lists.TryGetList(LogListName);             if (list == null)             {                 var listGuid = web.Lists.Add(LogListName, "Logging for the masses", SPListTemplateType.GenericList);                 list = web.Lists[listGuid];                 list.Title = LogListTitle;                 list.Update();                 list.Fields.Add(Category, SPFieldType.Text, false);                 var stringColl = new StringCollection();                 stringColl.AddRange(new[]{Error, Information, Verbose});                 list.Fields.Add(Severity, SPFieldType.Choice, true, false, stringColl);                 ModifyDefaultView(list);             }         }Should be self explanatory, but: only create the list if it does not already exist (d'oh). Best practice: create it with a Url-friendly name, and, if necessary, give it a better title. ...because otherwise you'll have to look for a list with a name like "Simple_x0020_Log". I've added a couple of fields; a field for category, and a 'severity'. Both to make it easier to find relevant log messages. Notice that I don't have to call list.Update() after adding the fields - this would cause a nasty error (something along the lines of "List locked by another user"). The function for deleting the log is exactly as onerous as you'd expect:         public void DeleteLog(SPWeb web)         {             var list = web.Lists.TryGetList(LogListTitle);             if (list != null)             {                 list.Delete();             }         } So! "All" that remains is to log. Also known as adding items to a list. Lots of different methods with different signatures end up calling the same function. For example, LogVerbose(web, message) calls LogVerbose(web, null, message) which again calls another method which calls: private static void Log(SPWeb web, string category, string severity, string textformat, params object[] texts)         {             if (web != null)             {                 var list = web.Lists.TryGetList(LogListTitle);                 if (list != null)                 {                     var item = list.AddItem(); // NOTE! NOT list.Items.Add… just don't, mkay?                     var text = string.Format(textformat, texts);                     if (text.Length > 255) // because the title field only holds so many chars. Sigh.                         text = text.Substring(0, 254);                     item[SPBuiltInFieldId.Title] = text;                     item[Degree] = severity;                     item[Category] = category;                     item.Update();                 }             } // omitted: Also log to SharePoint log.         } By adding a params parameter I can call it as if I was doing a Console.WriteLine: LogVerbose(web, "demo", "{0} {1}{2}", "hello", "world", '!'); Ok, that was a silly example, a better one might be: LogError(web, LogCategory, "Exception caught when updating {0}. exception: {1}", listItem.Title, ex); For performance reasons I use list.AddItem rather than list.Items.Add. For completeness' sake, let us include the "ModifyDefaultView" function that I deliberately skipped earlier.         private void ModifyDefaultView(SPList list)         {             // Add fields to default view             var defaultView = list.DefaultView;             var exists = defaultView.ViewFields.Cast<string>().Any(field => String.CompareOrdinal(field, Severity) == 0);               if (!exists)             {                 var field = list.Fields.GetFieldByInternalName(Severity);                 if (field != null)                     defaultView.ViewFields.Add(field);                 field = list.Fields.GetFieldByInternalName(Category);                 if (field != null)                     defaultView.ViewFields.Add(field);                 defaultView.Update();                   var sortDoc = new XmlDocument();                 sortDoc.LoadXml(string.Format("<Query>{0}</Query>", defaultView.Query));                 var orderBy = (XmlElement) sortDoc.SelectSingleNode("//OrderBy");                 if (orderBy != null && sortDoc.DocumentElement != null)                     sortDoc.DocumentElement.RemoveChild(orderBy);                 orderBy = sortDoc.CreateElement("OrderBy");                 sortDoc.DocumentElement.AppendChild(orderBy);                 field = list.Fields[SPBuiltInFieldId.Modified];                 var fieldRef = sortDoc.CreateElement("FieldRef");                 fieldRef.SetAttribute("Name", field.InternalName);                 fieldRef.SetAttribute("Ascending", "FALSE");                 orderBy.AppendChild(fieldRef);                   fieldRef = sortDoc.CreateElement("FieldRef");                 field = list.Fields[SPBuiltInFieldId.ID];                 fieldRef.SetAttribute("Name", field.InternalName);                 fieldRef.SetAttribute("Ascending", "FALSE");                 orderBy.AppendChild(fieldRef);                 defaultView.Query = sortDoc.DocumentElement.InnerXml;                 //defaultView.Query = "<OrderBy><FieldRef Name='Modified' Ascending='FALSE' /><FieldRef Name='ID' Ascending='FALSE' /></OrderBy>";                 defaultView.Update();             }         } First two lines are easy - see if the default view includes the "Severity" column. If it does - quit; our job here is done.Adding "severity" and "Category" to the view is not exactly rocket science. But then? Then we build the sort order query. Through XML. The lines are numerous, but boring. All to achieve the CAML query which is commented out. The major benefit of using the dom to build XML, is that you may get compile time errors for spelling mistakes. I say 'may', because although the compiler will not let you forget to close a tag, it will cheerfully let you spell "Name" as "Naem". Whichever you prefer, at the end of the day the view will sort by modified date and ID, both descending. I added the ID as there may be several items with the same time stamp. So! Simple logging to a list, with sensible a view, and with normal functionality for creating your own filterings. I should probably have added some more views in code, ready filtered for "only errors", "errors and warnings" etc. And it would be nice to block verbose logging completely, but I'm not happy with the alternatives. (yetanotherfeature or an admin page seem like overkill - perhaps just removing it as one of the choices, and not log if it isn't there?) Before you comment - yes, try-catches have been removed for clarity. There is nothing worse than having a logging function that breaks your site!

    Read the article

  • java: List wrapper where get()/set() is allowed but add/remove is not

    - by Jason S
    I need to wrap a List<T> with some class that allows calls to set/get but does not allow add/remove calls, so that the list remains "stuck" at a fixed length. I think I have a thin wrapper class (below) that will work, but I'm not 100% positive. Did I miss anything obvious? import java.util.Collection; import java.util.Iterator; import java.util.List; import java.util.ListIterator; class RestrictedListWrapper<T> implements List<T> { static <T> T fail() throws UnsupportedOperationException { throw new UnsupportedOperationException(); } static private class IteratorWrapper<T> implements ListIterator<T> { final private ListIterator<T> iter; private IteratorWrapper(ListIterator<T> iter) { this.iter = iter; } static public <T> RestrictedListWrapper.IteratorWrapper<T> wrap(ListIterator<T> target) { return new RestrictedListWrapper.IteratorWrapper<T>(target); } @Override public void add(T e) { fail(); } @Override public boolean hasNext() { return this.iter.hasNext(); } @Override public boolean hasPrevious() { return this.iter.hasPrevious(); } @Override public T next() { return this.iter.next(); } @Override public int nextIndex() { return this.iter.nextIndex(); } @Override public T previous() { return this.iter.previous(); } @Override public int previousIndex() { return this.iter.previousIndex(); } @Override public void remove() { fail(); } @Override public void set(T e) { this.iter.set(e); } } final private List<T> list; private RestrictedListWrapper(List<T> list) { this.list = list; } static public <T> RestrictedListWrapper<T> wrap(List<T> target) { return new RestrictedListWrapper<T>(target); } @Override public boolean add(T arg0) { return fail(); } @Override public void add(int index, T element) { fail(); } @Override public boolean addAll(Collection<? extends T> arg0) { return fail(); } @Override public boolean addAll(int arg0, Collection<? extends T> arg1) { return fail(); } /** * clear() allows setting all members of the list to null */ @Override public void clear() { ListIterator<T> it = this.list.listIterator(); while (it.hasNext()) { it.set(null); it.next(); } } @Override public boolean contains(Object o) { return this.list.contains(o); } @Override public boolean containsAll(Collection<?> c) { return this.list.containsAll(c); } @Override public T get(int index) { return this.list.get(index); } @Override public int indexOf(Object o) { return this.list.indexOf(o); } @Override public boolean isEmpty() { return false; } @Override public Iterator<T> iterator() { return listIterator(); } @Override public int lastIndexOf(Object o) { return this.list.lastIndexOf(o); } @Override public ListIterator<T> listIterator() { return IteratorWrapper.wrap(this.list.listIterator()); } @Override public ListIterator<T> listIterator(int index) { return IteratorWrapper.wrap(this.list.listIterator(index)); } @Override public boolean remove(Object o) { return fail(); } @Override public T remove(int index) { fail(); return fail(); } @Override public boolean removeAll(Collection<?> c) { return fail(); } @Override public boolean retainAll(Collection<?> c) { return fail(); } @Override public T set(int index, T element) { return this.list.set(index, element); } @Override public int size() { return this.list.size(); } @Override public List<T> subList(int fromIndex, int toIndex) { return new RestrictedListWrapper<T>(this.list.subList(fromIndex, toIndex)); } @Override public Object[] toArray() { return this.list.toArray(); } @Override public <T> T[] toArray(T[] a) { return this.list.toArray(a); } }

    Read the article

  • Using a "white list" for extracting terms for Text Mining, Part 2

    - by [email protected]
    In my last post, we set the groundwork for extracting specific tokens from a white list using a CTXRULE index. In this post, we will populate a table with the extracted tokens and produce a case table suitable for clustering with Oracle Data Mining. Our corpus of documents will be stored in a database table that is defined as create table documents(id NUMBER, text VARCHAR2(4000)); However, any suitable Oracle Text-accepted data type can be used for the text. We then create a table to contain the extracted tokens. The id column contains the unique identifier (or case id) of the document. The token column contains the extracted token. Note that a given document many have many tokens, so there will be one row per token for a given document. create table extracted_tokens (id NUMBER, token VARCHAR2(4000)); The next step is to iterate over the documents and extract the matching tokens using the index and insert them into our token table. We use the MATCHES function for matching the query_string from my_thesaurus_rules with the text. DECLARE     cursor c2 is       select id, text       from documents; BEGIN     for r_c2 in c2 loop        insert into extracted_tokens          select r_c2.id id, main_term token          from my_thesaurus_rules          where matches(query_string,                        r_c2.text)>0;     end loop; END; Now that we have the tokens, we can compute the term frequency - inverse document frequency (TF-IDF) for each token of each document. create table extracted_tokens_tfidf as   with num_docs as (select count(distinct id) doc_cnt                     from extracted_tokens),        tf       as (select a.id, a.token,                            a.token_cnt/b.num_tokens token_freq                     from                        (select id, token, count(*) token_cnt                        from extracted_tokens                        group by id, token) a,                       (select id, count(*) num_tokens                        from extracted_tokens                        group by id) b                     where a.id=b.id),        doc_freq as (select token, count(*) overall_token_cnt                     from extracted_tokens                     group by token)   select tf.id, tf.token,          token_freq * ln(doc_cnt/df.overall_token_cnt) tf_idf   from num_docs,        tf,        doc_freq df   where df.token=tf.token; From the WITH clause, the num_docs query simply counts the number of documents in the corpus. The tf query computes the term (token) frequency by computing the number of times each token appears in a document and divides that by the number of tokens found in the document. The doc_req query counts the number of times the token appears overall in the corpus. In the SELECT clause, we compute the tf_idf. Next, we create the nested table required to produce one record per case, where a case corresponds to an individual document. Here, we COLLECT all the tokens for a given document into the nested column extracted_tokens_tfidf_1. CREATE TABLE extracted_tokens_tfidf_nt              NESTED TABLE extracted_tokens_tfidf_1                  STORE AS extracted_tokens_tfidf_tab AS              select id,                     cast(collect(DM_NESTED_NUMERICAL(token,tf_idf)) as DM_NESTED_NUMERICALS) extracted_tokens_tfidf_1              from extracted_tokens_tfidf              group by id;   To build the clustering model, we create a settings table and then insert the various settings. Most notable are the number of clusters (20), using cosine distance which is better for text, turning off auto data preparation since the values are ready for mining, the number of iterations (20) to get a better model, and the split criterion of size for clusters that are roughly balanced in number of cases assigned. CREATE TABLE km_settings (setting_name  VARCHAR2(30), setting_value VARCHAR2(30)); BEGIN  INSERT INTO km_settings (setting_name, setting_value) VALUES     VALUES (dbms_data_mining.clus_num_clusters, 20);  INSERT INTO km_settings (setting_name, setting_value)     VALUES (dbms_data_mining.kmns_distance, dbms_data_mining.kmns_cosine);   INSERT INTO km_settings (setting_name, setting_value) VALUES     VALUES (dbms_data_mining.prep_auto,dbms_data_mining.prep_auto_off);   INSERT INTO km_settings (setting_name, setting_value) VALUES     VALUES (dbms_data_mining.kmns_iterations,20);   INSERT INTO km_settings (setting_name, setting_value) VALUES     VALUES (dbms_data_mining.kmns_split_criterion,dbms_data_mining.kmns_size);   COMMIT; END; With this in place, we can now build the clustering model. BEGIN     DBMS_DATA_MINING.CREATE_MODEL(     model_name          => 'TEXT_CLUSTERING_MODEL',     mining_function     => dbms_data_mining.clustering,     data_table_name     => 'extracted_tokens_tfidf_nt',     case_id_column_name => 'id',     settings_table_name => 'km_settings'); END;To generate cluster names from this model, check out my earlier post on that topic.

    Read the article

< Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >